NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Kuhn CM, Koob GF, editors. Advances in the Neuroscience of Addiction. 2nd edition. Boca Raton (FL): CRC Press/Taylor & Francis; 2010.

Cover of Advances in the Neuroscience of Addiction

Advances in the Neuroscience of Addiction. 2nd edition.

Show details

Chapter 2Application of Chronic Extracellular Recording Method to Studies of Cocaine Self-Administration

Method and Progress

, , and .

2.1. CHAPTER OVERVIEW

To further the investigation of the neurobiology of drug addiction, researchers have integrated chronic extracellular recording and intravenous drug self-administration procedures. The technique allows researchers to characterize the discharge patterns of single neurons during drug-directed instrumental behavior. This method is the topic of the present chapter. The first sections of the chapter overview the method and discuss the rationale, advantages, and disadvantages of the technique. These sections also describe some of the experimental approaches that are required for certain applications of the recording procedure (see Sections 2.2-2.4). The remainder of the chapter will review three areas of research, as follows: (1) firing patterns of NAc neurons during drug-directed behavior (Sections 2.5-2.6); (2) acute effects of cocaine on NAc activity (Section 2.7); and (3) effects of repeated cocaine self-administration on NAc firing (Section 2.8). The research exemplifies some of the experimental methods discussed in earlier parts of the chapter, as well as some of the novel research opportunities that are afforded by the technique. The chapter also includes a discussion of future directions. A detailed description of technical aspects of the method is provided in an appendix.

2.2. BASIC METHOD

The intravenous drug self-administration model is a well-validated model of human drug self-administration (Johanson and Balster 1978; Griffiths et al. 1980; Stewart et al. 1984; O'Brien et al. 1998). Under appropriate conditions of drug exposure, the method is also expected to serve as a valid model of drug addiction (Shalev et al. 2002; Shaham et al. 2003; Panlilio and Goldberg 2007). In the simplest case, animals are implanted with an intravenous catheter and then trained to engage in operant behavior, typically a lever press response, according to a Fixed-Ratio 1 (FR1) schedule of reinforcement. Each press of the lever is followed by a single infusion of drug, which is typically paired with a conditioned reinforcer such as a light or a tone. Although the experimenter controls the drug concentration and volume of each infusion, animals control the timing of infusions. Within the confines of drug availability, humans also control their own drug exposure. Interestingly the patterns of intake exhibited by animals in the self-administration paradigm are quite similar to those of humans. Drug exposure is an important determinant of both acute and chronic drug effects. Use of the self-administration procedure increases the likelihood that animals are exposed to levels and patterns of drug exposure that are consistent with those of drug addicts. Researchers using the self-administration paradigm are thus more likely to characterize drug effects that are relevant to drug addiction, as compared to experimenter-determined injections.

If exposed to short daily sessions and moderate drug doses, animals exhibit stable rates of intake across many days. If, on the other hand, animals are exposed to high doses of drug and/or long daily sessions, animals exhibit escalation of drug intake across repeated self-administration sessions, as well as other behaviors that are considered addiction-like (Koob et al. 1989; Ahmed and Koob 1998; Koob and Le Moal 2001, 2008; Deroche-Gamonet et al. 2004; Vanderschuren and Everitt 2004). The emergence of these addiction-like behaviors provides additional evidence of the validity of the self-administration paradigm as a model for studies of drug addiction. Moreover, brain areas implicated in mediating drug self-administration in animals are consistent with regions implicated in drug reward and addiction in humans (Grant et al. 1996; Breiter et al. 1997; Childress et al. 1999; Volkow et al. 1999). Identification of the neural mechanisms that mediate drug self-administration behavior is expected to contribute to delineating causes of drug addiction.

In chronic extracellular recording studies of drug self-administration (referred to herein as chronic behaving animal recording studies), animals are chronically implanted with either a bundle or an array of insulated microwires (e.g., 50 μm dia stainless steel wires with Teflon-coating) (Figure 2.1). During recording sessions, a headstage containing electrical contacts to the microwires is connected via a flexible cable to a recording system. Animals can move freely and exhibit typical operant behavior. During recording sessions, voltage signals from each microwire pass through an amplification and filtering system (Figures 2.2-2.3). Signals that meet preselected criteria (e.g., amplitude criteria) are time stamped and stored for subsequent offline analysis. After the experiment, analysis software is used to discriminate voltage signals that correspond to action potentials of individual neurons (Figure 2.4). Histogram analyses are then used to test for changes in firing rate of single neurons in relation to particular events (Figure 2.5). A more detailed description of the technique is included in an appendix at the end of this chapter. Additional details, as well as a number of equipment photographs, are included in (Peoples 2003).

FIGURE 2.1. Microwire array.

FIGURE 2.1

Microwire array. Photograph of the tip of a 16-wire microwire array. The array is 2mm in width and 1mm in height. Photograph shows wire tips emerging from polyethylene glycol, which is necessary for maintaining the configuration of the wires but is slowly (more...)

FIGURE 2.2. Rat engaged in intravenous cocaine self-administration during a chronic extracellular recording session.

FIGURE 2.2

Rat engaged in intravenous cocaine self-administration during a chronic extracellular recording session. Photograph shows a rat pressing the operant lever during a cocaine self-administration session. The headstage of the animal is connected to an electronic (more...)

FIGURE 2.3. Basic electrophysiological hardware setup.

FIGURE 2.3

Basic electrophysiological hardware setup. Med Associates operant hardware and software are integrated with the Plexon electrophysiological recording system. Microwire signals pass through a headset amplifier mounted on the rat's head to a differential (more...)

FIGURE 2.4. Example of extracellular recordings of single-neuron discharges.

FIGURE 2.4

Example of extracellular recordings of single-neuron discharges. A. Scatterplot of the first two principal components (PC1 and PC2) of all large-amplitude electrical events recorded from one microwire during 15 minutes of an operant session. Events with (more...)

FIGURE 2.5. Example of a raster and histogram display for a single neuron.

FIGURE 2.5

Example of a raster and histogram display for a single neuron. Top: A raster display marks the occurrence of each discharge of a particular neuron for 15 cocaine-reinforced lever presses. Each tick corresponds to a single discharge. Each row shows all (more...)

The occurrence of action potentials can be tracked with a 1 msec resolution. The recordings are highly stable such that the activity of a single neuron can be followed for many hours and sometimes for multiple days (Peoples et al. 1999a; Janak 2002). It is thus possible to characterize changes in neuronal firing that occur time-locked to rapid discrete events, such as the cocaine-reinforced operant response (i.e., msec to sec time frame). One can also concurrently evaluate slow and long-lasting changes in firing that might be associated with changes in either drug exposure or the motivational state of the animal (i.e., time frame of min to hrs).

With histological procedures, the location of recording wires can be identified with a resolution of ≈100 μm) (Uzwiak et al. 1997). It is thus possible to use the procedure to test for differences in single neuron activity across relatively small sub-regions of a given structure (Uzwiak et al. 1997; Hollander and Carelli 2005; Peoples et al. 2007b).

2.3. RATIONALE, ADVANTAGES, AND DISADVANTAGES OF THE METHOD

2.3.1. Importance of Electrophysiological Investigations for Identifying Causes and Treatments of Addiction

Drug addiction is a progressive and chronic disorder characterized by the emergence of compulsive drug seeking and taking. It is also associated with a concomitant weakening of other motivated behaviors (Gawin and Kleber 1986; Weddington et al. 1990; American Psychiatric Association 1994). It is hypothesized that the maladaptive behavior that characterizes addiction is mediated by long-lasting drug-induced neuroadaptations within neuronal circuits that control motivated behavior (Berke and Hyman 2000; Everitt and Wolf 2002; Robinson and Berridge 2003; Kalivas 2004; Thomas et al. 2008). Identifying the neuroadaptations and the drug actions that mediate the adaptations are necessary steps for understanding the causes of addiction as well as for the development of effective pharmacotherapies.

Drug-induced adaptations involve changes in protein function in individual neurons, which can impact the excitability and activity of the individual neurons. Changes in single neuron firing can alter the activity of neuronal circuits and lead to changes in behavior. Understanding the specific alterations in protein function is a critical aspect of identifying drug-induced adaptations that might contribute to drug addiction. However, it is also critical to identify the effects of drug on neuronal activity.

Addictive drugs induce a plethora of changes in protein synthesis and function (Nestler 2001; Self et al. 2004; Hope et al. 2005; Lu et al. 2005; Ben-Shahar et al. 2007; Zhang et al. 2007; Kalivas and O'Brien 2008). Understanding the impact of drug exposure on the excitability and activity of individual neurons can help to differentiate among changes in protein function that might be of greatest functional consequence. Additionally, it is possible that mechanisms that mediate addiction involve changes in the activity of neuronal microcircuits, as well as regional circuits. Knowledge of drug-induced changes in single neuron and circuit function, combined with an understanding of the drug-induced changes in protein function that underlie the neurophysiological plasticity, is necessary for a full understanding of mechanisms that mediate the development of addiction. The multilevel characterization of drug-induced plasticity is also expected to identify a range of potential targets for pharmacotherapy. These targets may involve primary sites of drug action, but could also involve secondary sites, that might be targeted to compensate or reverse drug-induced changes in protein function, neural excitability, or circuit function.

2.3.2. Characterization of Acute and Chronic Drug Effects on Single Neuron Activity

Identifying drug-induced neurophysiological plasticity and the mechanisms that mediate the plasticity involves characterization of neurophysiological changes in neuronal function that occur as a consequence of repeated drug exposure. However, neuroplasticity is importantly influenced by acute drug actions. Acute pharmacological actions can influence neuroadaptative processes in three ways. First, acute drug actions stimulate receptors and signal transduction pathways. These actions lead to the onset of adaptations in the affected neurons. Second, acute drug actions impact neuron firing rates during the period of drug exposure. These changes in activity can impact receptor-mediated changes in signal transduction pathways in the affected neuron and thereby influence the occurrence of adaptations (Wolf et al. 2004). Third, acute drug-induced changes in neural activity can alter afferent input into other circuits and potentially influence adaptive processes ongoing in those circuits (e.g., drug could alter inputs to memory circuits). A complete understanding of drug-induced neuroadaptations thus depends on characterization of both acute and chronic drug effects.

2.3.3. Studies of Acute and Chronic Drug Effects: Advantages and Disadvantages of Different Electrophysiological Approaches

Many studies of acute and chronic effects of cocaine on neuronal activity have employed either slice recording preparations or anesthetized animals. These electrophysiological methods have a number of advantages. The methods can be used to characterize specific biochemical and molecular mechanisms that mediate drug-induced changes in neural activity (White and Kalivas 1998; Zhang et al. 2002; Hu et al. 2004, 2005; Nasif et al. 2005).

Despite the advantages of the anesthetized animal and slice procedures, the techniques have certain limitations. Drug effects on single neuron activity can depend on multiple factors. These include but are not limited to the state of depolarization, the firing rate of a neuron, and the identity of afferent input to a neuron at the time of drug exposure (O'Donnell et al. 1999; Floresco et al. 2003; O'Donnell 2003; West et al. 2003). There is a potential for marked differences in the state of these variables between behaving animals and either the slice or the anesthetized animal preparations. In both the slice and anesthetized animal preparations, individual neurons can be cut off from normal afferent input. Anesthesia also directly impacts the response of neurons to afferent input (Mereu et al. 1983; Yoshimura et al. 1985; Fink-Jensen et al. 1994; Antkowiak and Helfrich-Forster 1998; Antkowiak, 1999). Moreover, particular behaviors are associated with unique patterns of afferent input to individual neurons, which are necessarily absent in the slice and anesthetized animal preparations. Given these differences and the condition-dependent nature of drug actions, it is possible that certain acute and chronic drug effects that occur in the slice and anesthetized animal preparations are not representative of those that occur in the behaving animal. It is also possible that certain effects of drug that occur in the behaving animal and that contribute to addiction are absent in the slice and anesthetized animal preparations.

The potential for different findings between recordings in the slice and anesthetized animal preparations, and the recordings in behaving animals is emphasized by the following: (1) previous demonstrations of such differences (Deadwyler 1986; West et al. 1997; Moxon 1999), and (2) previous observations of behavior-dependent drug effects on various neural substrates (Smith et al. 1980,1982; Hemby et al. 1997). Additionally, a number of extracellular recording studies during periods of ongoing cocaine seeking and self-administration show that there are differences in acute and chronic cocaine effects that are observed between the behaving animal recording studies and the slice and anesthetized animal preparations (Peoples and Cavanaugh 2003; Peoples et al. 2004,2007b) (see Sections 2.7-2.8).

Based on the above observations, recordings in animals could be essential for identifying the acute and chronic drug effects on neuronal signaling that mediate the development of addiction.

Electrophysiological studies in behaving animals also provide an additional unique research opportunity. The chronic-behaving animal recording methods can be used to conduct fine-grained correlational analyses of the activity of individual neurons and ongoing sensory and behavioral events. This measurement capacity provides a unique opportunity for testing functional hypotheses of drug addiction (see Sections 2.4.2 and 2.8.4).

Despite the advantages of the chronic-behaving animal recording technique, use of the method in investigations of acute and chronic drug effects is associated with challenges and limitations. In behaving animals, a change in the activity of a single neuron that occurs in response to acute drug exposure can reflect either of the following: (1) pharmacological actions on the recorded neuron, or primary afferent input to the neuron, and (2) an effect of a nonpharmacological factor on the firing of the recorded neuron. In regard to the latter, drug actions in nonrecorded regions can lead to changes in behavior, sensory processes, and motivational processes. Recorded neurons potentially receive afferent input related to those behaviors and processes (i.e., feedback). If this is the case, changes in activity of a neuron during drug exposure can be entirely due to the changes in behavior, sensory, or motivational processes rather than to an effect of drug on the recorded neurons. Similarly, a change in neural activity that is observed in association with repeated exposure to drug self-administration potentially corresponds to either of the following: (1) drug-induced neuroadaptations or (2) normal plasticity associated with repeated exposure to nonpharmacological aspects of the experimental procedures, such as operant conditioning. It is necessary to differentiate the relative contribution of pharmacological and nonpharmacological variables for the findings of the studies to be useful in identifying and investigating drug effects.

With appropriate controls it is possible to identify pharmacologically mediated changes in neural activity in behaving animals. However, characterization of pharmacological mechanisms that contribute to addiction involves not only identification of drug effects but investigation of the specific drug actions that mediate those effects. There are a number of procedures that are typically combined with anesthetized and slice electrophysiological recording techniques to characterize drug actions. Unfortunately, the methods are not readily incorporated with the chronic recording method.

One of these methods involves the use of dye and tracing techniques can be used to identify the specific type of neuron from which neural recordings are made: moreover, the techniques can be used to identify the specific afferent inputs and cellular proteins that are either involved in or affected by drug actions. Another method for investigating the mechanisms of drug action is to use iontophoretic and microinjection techniques to make local pharmacological manipulations of potential sites of drug action. Application of these techniques is difficult in behaving animals but it is especially complicated by the chronic implantation of microwires. The chronic implant of the fine wires prevents mechanical displacement of the wires in brain and is necessary for obtaining the long-duration, stable recordings in the behaving animal. Animals can be very active during drug self-administration sessions and recordings sometimes have to be extended for hours in order to obtain a sufficient sample of neural responses to task-related events such as operant behavior, which can occur at a slow rate during drug self-administration (e.g., at an intermediate dose of cocaine, animals might press the lever once every 6–8 min). With technical development efforts, it is most likely possible to integrate some of the techniques used in other recording studies to the chronic recordings in behaving animals, or to develop alternative methods of investigating mechanisms of drug action in the behaving animal. At the moment, however, opportunities for that type of research are quite limited. Possible avenues for improving this situation are discussed in Future Directions (Section 2.9).

2.3.4. An Integrated Electrophysiological Approach

Given the relative strengths and weaknesses of the different electrophysiological methods, complementary application of the techniques, and consideration of findings from those studies as a whole, will be important to understanding the drug effects that contribute to drug addiction. Slice and anesthetized animal methods can be used to identify drug effects and to characterize mechanisms of drug action. Recordings in behaving animals can be used to test for changes in neuronal activity that are predicted on the basis of the slice and anesthetized recordings, and to test whether the changes in neural activity are associated with the development of addiction-like patterns of behavior. The advantages that can be gained by such an integrative approach are demonstrated by the research described in the present chapter.

2.4. APPLICATION OF THE METHOD: DIFFERENTIATING PHARMACOLOGICAL AND NONPHARMACOLOGICAL EFFECTS AND TESTING ADDICTION HYPOTHESES

2.4.1. Methods for Differentiating between Pharmacological and Nonpharmacological Effects on Neural Activity

2.4.1.1. Acute Drug Effects

Numerous methods can be used to differentiate between a pharmacological effect and a behavioral feedback effect on single neuron activity. One simple strategy for making the differentiation is to compare the time course of the changes in firing that occur after a pharmacological manipulation to the time course of changes in behavior. A dissociation between the time courses shows that the firing rate changes cannot be attributed to behavioral changes. A related strategy is to compare the dose response curves for drug effects on firing and particular behaviors. A difference between the dose response curves for changes in neural activity and behavior can indicate that the two dependent measures are unrelated. A third strategy is to evaluate the effect of manipulating the behavior on neural activity under drug-free conditions. If manipulation of the behavior does not engender a change in firing that mimics the effect of acute drug exposure, then the change in firing observed during drug exposure cannot be attributed to changes in that behavior. A fourth control for the effects of behavioral feedback is a behavioral clamp. The goal of the behavioral clamp is to make comparisons of different drug conditions across periods in which behavior is constant. If the originally observed drug-correlated change in firing is unaltered by the behavioral clamp, the neural change cannot be attributed to the behavior (Ranck et al. 1983). A detailed example of this control is included in Section 2.7.

There are several approaches to behavioral clamping. In some cases certain spontaneous behaviors occur during all drug exposure conditions, albeit with different frequencies. If one were concerned that the different frequency with which the behavior occurs was mediating changes in activity of the recorded neuron, one could limit comparisons of firing rate to periods in which that particular behavior occurs, for example, sample firing rate only during periods in which the animal is engaged in locomotion. Making comparisons of firing rates during these periods holds the impact of that behavior constant across the different pharmacological conditions. Another behavioral clamp strategy is to condition animals to engage in a particular behavior under the different drug exposure conditions and to limit firing-rate comparisons to the periods of the conditioned behavior.

In addition to changing behavior per se (e.g., inducing increases in locomotion or stereotypy), the acute effects of addictive drugs have reinforcing properties. Within a limited range, increments in doses increase the strength of these reinforcing properties, consistent with effects of increasing the magnitude of natural rewards (Pickens and Thompson 1968; Johanson and Schuster 1975; Richardson and Roberts 1996; Arroyo et al. 1998; Olmstead et al. 2000; Mantsch et al. 2001; Peltier et al. 2001). The changes in reinforcer magnitude can in turn impact incentive processes and operant behavior. In addition to the primary reinforcing properties, acute effects of addictive drugs, within a limited dose range, can increase the invigorating and conditioned reinforcing properties of conditioned reward-associated stimuli on instrumental behavior (Robbins 1978; Wyvell and Berridge 2000), and can additionally increase Pavlovian conditioned approach behavior (Wan and Peoples 2008). Thus, if one is recording neurons in a region that is potentially involved in tracking aspects of these motivational processes, it is necessary to consider whether changes in neural activity reflect a neuronal response to the altered motivational process, or a more direct effect of drug on the recorded neuron. The approaches that one can use to differentiate between a change in “motivational feedback” and a more direct pharmacological effect are comparable to those described above in reference to behavioral feedback.

2.4.1.2. Chronic Drug Effects

Drug addiction symptoms develop progressively across a history of repeated drug use. Some addicts never attain a significant period of extended abstinence. Others may cycle between periods of active drug use, extended periods of abstinence, and relapse to active drug use. A full understanding of neuroadaptations underlying addiction may depend on characterizing neural activity across these repeated phases of drug exposure, extended abstinence, and re-exposure. Such studies will involve making comparisons of neural activity at different time points across a history of repeated daily self-administration sessions, and in some cases across phases of active drug use, drug abstinence, and drug re-exposure.

With chronically implanted wires, it is possible to follow the activity of individual neurons for a period of days to perhaps two weeks. However, most neurons cannot be reliably recorded for more than several days (Peoples et al. 1999a; Janak 2003). Moreover, studies that aim to characterize effects of cycles of drug self-administration, abstinence, and re-exposure need to maintain recordings for periods of months. A more practical alternative for characterizing chronic drug effects in behaving animal recording studies is to follow a between-group approach. That is, one can record separate groups of neurons at the different time points in the drug self-administration history and then test for an effect of repeated drug exposure by comparing average firing of neurons across the different time points. Though one can look at overall, average firing of the population at each time point, it is useful to take advantage of the spatial resolution of the technique and to subtype recorded neurons into groups that exhibit particular types of firing patterns during the self-administration sessions. For example, at each time point, one can identify neurons that show phasic responses to conditioned drug-associated stimuli and then compare the average firing of those different groups of cue-responsive neurons to test for an effect of repeated drug sessions, and so forth. This type of approach has been used successfully in several studies that will be discussed later in the chapter (see Section 2.8).

As is the case for acute drug effects, a challenge in behaving animal studies of chronic drug effects is to differentiate changes in firing that are of a primary pharmacological origin from those that are not. Changes in neural activity across repeated daily self-administration sessions can reflect cocaine-induced neuroadaptations in the recorded neuron. But, they could also reflect normal neural responses to repeated exposure to certain experimental treatments (i.e., neural correlates of learning). To be useful in identifying cocaine-induced neuroadaptations, it is necessary for behaving animal studies to differentiate between these possibilities. There are a number of approaches that can be employed. First, changes in neural activity that are based on processes such as habituation or associative learning should be situation-dependent, showing rapid changes in response to manipulations of various nonpharmacological aspects of the experimental treatment (e.g., the response-reinforcer condition or the operant). Changes in neural activity that reflect long-lasting drug-induced adaptations are expected to be relatively insensitive to similar manipulations. A second and more convincing method of differentiating drug-induced adaptation from experiential effects on neural activity is to compare the changes in firing observed in animals exposed to similar treatments but trained to self-administer a natural reward instead of a drug reward. Given a carefully controlled study, the selective occurrence of neural changes in the drug-exposed animals supports a pharmacological interpretation of the drug data. Examples of these approaches are described later in the chapter (see Section 2.8).

2.4.2. Testing Drug Addiction Hypotheses

Drug addiction is characterized by disruptions in motivated behavior. Individuals exhibit compulsive and uncontrollable drug-seeking and taking, despite knowledge of adverse consequences. Individuals exhibit a weakened ability to use information about adverse consequences to regulate instrumental behavior and a narrowing of behavioral repertoire (i.e., decrease in alternative motivated behaviors). Consistent with these characteristics, addiction is hypothesized to be caused by drug-induced alterations in neural processes that normally regulate motivated behavior.

A number of researchers have put forth specific hypotheses. For example, the incentive sensitization hypothesis postulates that neuroadaptations alter mechanisms that attribute incentive salience to rewards and conditioned reward-associated stimuli. The drug-induced abnormalities are hypothesized to lead to an abnormally strong response of those mechanisms to drug and drug-associated cues (Robinson and Berridge 1993; Koob and Le Moal 1997; Koob and Nestler 1997). The homeostatic dysregulation hypothesis postulates that neuronal mechanisms that regulate affective state and hedonic responses to primary reward are altered so that the baseline affective state of an individual becomes highly negative. Moreover, the hedonic impact of natural rewards is no longer sufficient to normalize the affective state of individuals, though the direct effects of incrementing doses of drug are sufficient to do so (Koob and Le Moal 2001). The stimulus-response (S-R) learning hypothesis proposes that drug-induced adaptations lead to abnormalities in neuronal mechanisms that regulate habits, which are automatic behavioral responses induced by conditioned stimuli. The differential neuroplasticity hypothesis proposes that neurons that regulate natural reward-directed behavior and neurons that facilitate the drug-directed behavior undergo different, activity-dependent neuroadaptations. The differential neuroadaptations tend to weaken neural signaling related to natural reward and to strengthen signaling related to drug, which contributes to corresponding changes in behavior (Peoples and Cavanaugh 2003; Peoples et al. 2004, 2007a, 2007b). In addition to specifying functional disruptions, most hypotheses put forth specific proposals about the underlying brain regions.

All of the addiction hypotheses propose that repeated drug exposure leads to particular changes in neural responses associated with certain stimuli and behavior. Moreover these neuronal changes are expected to be associated with the emergence of addiction-like patterns of behavior. Electrophysiological recordings in behaving animals can be used to test these hypotheses. For example, the incentive sensitization hypothesis postulates that repeated exposure to addictive drugs will lead to increased incentive-related neuronal responses to drug-associated conditioned stimuli in the mesoaccumbal dopamine (DA) pathway. The homeostatic dysregulation hypothesis might predict that repeated drug exposure will cause a lasting shift in activity of NAc neurons that track hedonic state (if such exist) and a decreased effect of natural rewards on those neurons in animals with extensive drug histories. Based on the S-R hypothesis, dorsal striatal neuron responses involved in regulating instrumental behaviors are expected to be amplified after repeated drug exposure; whereas ventral striatal responses involved in flexible behavior would be weakened. Based on the differential neuroplasticity hypothesis one would predict both increased responses to drug-related events (such as drug-associated cues) and decreased responses to natural reward-related events (such as cues associated with natural rewards). In all cases, as the predicted changes in firing emerge, addiction-like behaviors are also expected to develop.

Electrophysiological recordings in behaving animals can be used to test the addiction hypotheses. Specifically, the procedure can be used to (1) identify baseline neural responses that correspond to those implicated by a specific hypothesis; (2) test for predicted effects of repeated drug exposure on those neural responses; and (3) test whether those changes in neuronal responses are associated with the emergence of the predicted, addiction-like behaviors. One of the most challenging aspects of this work is to identify neural responses associated with specific types of information related to motivated behavior. There are several approaches that could be used. First, one can use paradigms that temporally dissociate particular processes and events. In this case, neural responses related to one or more of those processes and events are expected to show a response at some time points and not at others. Alternatively, the paradigm can be designed to manipulate individual processes, holding others constant. If a neural response reflects the manipulated variable, then one should observe predicted changes in that response (i.e., prevalence or magnitude of response).

Sophisticated examples of these approaches can be found in primate recording studies of DA and dopaminoceptive target regions such as the ventral and dorsal striatum (Schultz 2000). These studies aimed to identify the specific role of DA and dopaminoceptive neurons in motivated behavior. Early studies showed that the neurons respond to discriminative stimuli that predict reward availability. Discriminative stimuli carry multiple types of information and activate multiple processes, including but not limited to response-reinforcer associations, which facilitate the occurrence of a specific operant, and stimulus-reward associations, which might activate Pavlovian behaviors and visceral processes related to consumption. Single neuron responses theoretically could have corresponded to one or more of those associations and events. It was also possible that changes in firing time-locked to the discriminative stimulus reflected sensory properties of the cue. One strategy used to differentiate among these possibilities was to temporally dissociate the different types of information. Animals were trained to discriminate multiple cues, which were presented separately. Each cue provided a particular type of information (e.g., the particular operant response that was required to earn the reward during an upcoming trial or the particular reward that was available on a given trial) (Apicella et al. 1991; Cromwell and Schultz 2003). These cues were temporally spaced so that researchers could test for changes in neural responses to each of the cues. Some neurons responded to a single cue, and other neurons responded to multiple cues. With this approach it was possible to differentiate among the types of information to which changes in neuron firing corresponded.

To characterize potential behavioral correlates of firing patterns, researchers manipulated particular variables, holding others constant. For example, the researchers were interested in testing whether firing patterns during instrumental behavior were related to the specific behavior. To test this possibility animals were trained to emit either of two behavioral responses, depending on which of two discriminative stimuli were presented. In these experiments, neural responses were stable across the trials that required different operant behaviors, indicating that those responses were unrelated to the cue-response associations. A similar approach can be used to test whether a cue response reflects primary sensory processing or a response to the associative characteristic of the cue. For example, researchers have trained animals to discriminate cues, which signal different rewards, and then tested the effect of reversing the cue-reward relationship. In such experiments, cue responses that reflect sensory properties of the cue are unaffected by the reversal of the cue-reward relationship. On the other hand, neural responses that reflect the associative (i.e., cue-reward relationship) properties of the cue show changes in activity (Cromwell and Schultz 2003; Calu et al. 2007; Roesch et al. 2007; Stalnaker et al. 2007).

Similar approaches can be used to isolate neuronal responses relevant to particular drug addiction hypotheses. Thus far, recording studies related to addiction have employed some of the simplest possible procedures. This is in part related to the multiple challenges associated with a rat model of intravenous self-administration, including but not limited to the time required to condition rats on sophisticated behavioral procedures and the limited life span of the intravenous catheter. However, advances related to testing addiction hypotheses have been made and more sophisticated studies are possible.

2.5. FIRING PATTERNS DURING COCAINE TAKING

2.5.1. Overview of the Section

One focus of recording studies of cocaine-directed behavior has involved the cataloging of firing patterns that occur during FR1 cocaine self-administration sessions. Another has been investigating the nonpharmacological factors that contribute to those firing patterns. This research is of interest for at least two reasons. First, identifying firing patterns that are nonpharmacological facilitates identification of those firing patterns that are potentially pharmacological. Additionally, as we have already described, the recording method can be used to test current hypotheses regarding NAc mechanisms that contribute to the development and expression of drug addiction. This research requires isolating neural responses that are related to particular motivational events.

Much progress has been made in documenting firing patterns exhibited by NAc neurons in animals with limited drug exposure. Some experiments have also provided information about the events that are encoded by the firing patterns, though more work in this area is necessary. This section of the chapter reviews some of this research, which exemplifies numerous methods that can be used to differentiate among particular nonpharmacological determinants of firing patterns, as well as to differentiate between pharmacological and nonpharmacological underpinnings of the firing patterns.

2.5.2. Short-Duration Changes in Firing Time-Locked to the Reinforced Lever Press

2.5.2.1. Description of the Firing Patterns

Some of the first extracellular recordings conducted during FR1 cocaine self-administration sessions characterized phasic firing patterns time-locked to the operant behavior. These studies (Carelli et al. 1993; Chang et al. 1994; Peoples and West 1996; Uzwiak et al. 1997) showed that at intermediate dose of cocaine (e.g., 0.75 mg/ kg/inf), approximately 25%—40% of NAc neurons exhibit rapid phasic changes in firing time-locked to cocaine-reinforced lever presses (referred to herein as lever-press firing patterns). The lever-press firing patterns vary in time course, with the change in firing occurring primarily within a 9-sec period that brackets the reinforced press (i.e., 3-sec prepress to the 6-sec postpress) (Figure 2.6). We have subtyped these firing patterns into the following five groups: (1) exclusively prepress, (2) predominantly prepress, (3) symmetrically pre-post press, (3) predominantly postpress, and (4) exclusively postpress.

FIGURE 2.6. Rapid phasic changes in firing time-locked to the cocaine-reinforced lever press during an FR1 cocaine self-administration session.

FIGURE 2.6

Rapid phasic changes in firing time-locked to the cocaine-reinforced lever press during an FR1 cocaine self-administration session. Each histogram shows an individual neuron example of an excitatory lever-press firing pattern. In each of the three histograms, (more...)

Carelli and colleagues have described a “dual peak” firing pattern time-locked to the cocaine reinforced operant, which they have thus far not observed in studies of nondrug rewards. The pattern consists of two discrete phasic changes in firing, one prepress and one postpress. However, patterns similar to the dual peak response time-locked to the cocaine-reinforced operant have been described in primate studies of natural rewards (Schultz et al. 1992; Schultz 2000). It thus appears that the patterns are not unique to cocaine.

Phasic lever-press firing patterns have been observed under a broad range of experimental conditions but have been consistently reported as predominantly excitatory (Carelli et al. 1993; Chang et al. 1994; Uzwiak et al. 1997; Hollander and Carelli 2005; Peoples et al. 2007b). The firing patterns occur in both core and anterior portions of the shell but are nearly absent in the posterior portions of the shell (Uzwiak et al. 1997).

In addition to these rapid phasic firing patterns a small additional portion of neurons (≈30%) show longer duration (e.g., 30–60 sec) lever-press firing patterns (not shown). These neurons show either an increase prepress, an increase postpress, or a decrease postpress. Though the patterns are long in duration, they are punctuate with rapid onsets and offsets (Peoples and West 1996). The broad range of firing pattern time courses suggests that during cocaine self-administration, NAc neurons track multiple variables or combinations thereof during the seconds that precede and follow the reinforced operant response.

2.5.2.2. Determinants of the Firing Patterns

The offset of the “rapid” lever-press firing patterns precedes the expected latency for pharmacological actions postinfusion. There are numerous nonpharmacological events that occur within the seconds that lead up to and follow the reinforced lever press during typical FR1 cocaine self-administration sessions. These include, but are not limited to the following: approach to the lever, the lever-press operant, the conditioned reinforcer cues, and delivery of the primary reinforcer. The phasic firing patterns time-locked to the operant potentially encode information related to one or more of these events.

Studies were conducted to identify the events associated with the phasic firing patterns. These studies employed a number of analytic and experimental strategies. For example, researchers compared the time course of the firing patterns and the occurrence of specific behaviors during the seconds leading up to and following completion of the lever-press operant (Chang et al. 1994). In other studies, investigators dissociated particular events in time or manipulated particular events, while holding other nonpharmacological and pharmacological variables constant (Peoples et al. 1997; Carelli 2000). We will describe one example of the latter approach in this subsection.

In one study (Peoples et al. 1997), we tested for responses that were related to either the conditioned reinforcer or the instrumental behavior. Animals were trained on the FR1 cocaine self-administration session. We then conducted a multiphase recording session. At the beginning of the session, subjects self-administered cocaine infusions as usual. Thereafter, the session consisted of alternating phases in which infusions were either response contingent or response noncontingent. During each response-contingent phase, presentation of the conditioned cues and the associated cocaine infusion (5 or 15) occurred only when the rat depressed the lever. In each response-noncontingent phase, the presentation of the cue and infusion occurred in the absence of operant behavior. The infusion was presented with the exteroceptive light plus tone cue because proprioceptive or pump-related auditory cues might be associated with the onset of the infusion, which could accrue conditioned stimulus properties. The timing of the cue-infusion presentations followed the schedule of cue + infusion presentations self-administered by the rat during the preceding response-contingent phase. Lever presses during the noncontingent phase were nonreinforced (i.e., were not followed by either cues or cocaine infusions) (Figure 2.7).

FIGURE 2.7. Patterns of lever presses and calculated drug levels during the response-contingent and response-noncontingent cocaine phases.

FIGURE 2.7

Patterns of lever presses and calculated drug levels during the response-contingent and response-noncontingent cocaine phases. Patterns of behavior are shown for each of two animals during the successive phases of response-contingent, response-noncontingent, (more...)

The procedure controlled for multiple factors that could potentially influence NAc firing. Given that the schedules of contingent and noncontingent infusions were identical, drug level at the time of the infusion was held constant. Hence, the motivational state of the animal at the time of the cue and infusion was expected to be quite similar between the phases. Given the selective elimination of operant behavior during the seconds preceding the response-noncontingent cue presentations, firing patterns that were associated with the operant were expected to be present during the response-contingent phase but not during the response-noncontingent phase. Such firing patterns were also expected to occur when animals made nonreinforced operant responses. On the other hand, firing patterns that were associated with the conditioned cues were expected to persist across the response-contingent and response-noncontingent phase.

Of the 70 neurons recorded in the study, 29 showed a lever-press firing pattern during the contingent phases of the session. During the noncontingent infusion phase, all neurons (4/4) that showed a predominant prepress firing pattern time-locked to the response-contingent cue presentations did not show a change in firing time-locked to the response-noncontingent cue presentations. The same was true for almost all of the neurons (8/9) that showed a symmetrical firing pattern, and for about half of the neurons that showed either a predominantly (3/7) or exclusively (3/6) postpress firing pattern. For the other neurons (i.e., 10% of symmetrically responsive neurons and half of the predominantly and exclusively postpress responsive neurons), the neurons showed a phasic response time-locked to both response-contingent and response-noncontingent cue presentations, though the responses during the noncontingent phase tended to be diminished relative to that during the contingent phase (Figures 2.82.9). These findings show that predominantly and exclusively prepress firing patterns, almost all symmetrical lever-press firing patterns, and half of the predominantly and exclusively postpress firing patterns reflect neuronal responses associated with the operant. For the other neurons, the firing patterns reflect neuronal responses associated with both the operant and the cues. These findings are consistent with those of other studies (Carelli and Ijames 2000; Carelli and Wightman 2004).

FIGURE 2.8. Phasic firing time-locked to response-contingent cocaine infusions, response-noncontingent cocaine infusions, and unreinforced lever presses.

FIGURE 2.8

Phasic firing time-locked to response-contingent cocaine infusions, response-noncontingent cocaine infusions, and unreinforced lever presses. Each panel (A-D) shows the firing patterns of a single neuron. All neurons in the figure exhibited a phasic firing (more...)

FIGURE 2.9. Phasic firing time-locked to response-contingent cocaine infusions, response-noncontingent cocaine infusions, and unreinforced lever presses.

FIGURE 2.9

Phasic firing time-locked to response-contingent cocaine infusions, response-noncontingent cocaine infusions, and unreinforced lever presses. Both panels (A and B) show firing of a single neuron. Both neurons exhibited an exclusively postpress firing (more...)

Several studies showed that the lever-press firing patterns are specific for the expected reward. In these studies, NAc activity was recorded during multiphase sessions in which the reinforcer varied between phases and was either a natural reward (water or food) or cocaine (Carelli 2000). During these sessions most neurons that showed a phasic response time-locked to cocaine-reinforced lever presses did not show a phasic response time-locked to lever presses reinforced by the natural reward (Carelli et al. 2000). Similarly, other neurons that exhibited responses time-locked to lever presses reinforced by the natural reward rarely did the same when the operant was cocaine reinforced. In some of these experiments, the animals pressed the same lever during both phases, so the between-phase difference in firing could not have reflected either the movements associated with the operant or the spatial location of the manipulanda. These findings are consistent with the interpretation that the lever-press firing patterns are related in some way to expectation of a particular reward.

Additional studies are required to further specify the information encoded by the phasic firing patterns. However, available data on cocaine show that task-related neural activity during cocaine self-administration sessions is quite similar to that which occurs during behavior directed toward natural rewards. In both cases, NAc neurons exhibit phasic firing in association with stimulus events that engender an incentive state (e.g., cues that signal the availability of reward) and to behavioral events that are controlled by those states (Schultz et al. 1992; Lavoie and Mizumori 1994; Bowman et al. 1996; Hollerman et al. 1998; Cromwell and Schultz 2003). The firing patterns are exhibited by a relatively small subset of recorded neurons. The patterns involve predominantly increases in firing. The excitations are dissociable from both specific movements and physical properties of the cues and depend on reward expectation. The nature of the relation to reward expectation has been investigated in some detail in studies of natural rewards. Those studies show that the cue-associated firing patterns depend strongly on the predicted reward. Specifically, the magnitude of cue-locked phasic firing patterns varies depending on the following: (1) the reward type predicted by the cue (Hassani et al. 2001), (2) the magnitude of the predicted reward (Hollerman et al. 1998; Cromwell and Schultz 2003), and (3) the temporal proximity of the reward (Shidara et al. 1998). Whether the same is true for the cue-associated firing patterns during cocaine self-administration sessions remains to be determined.

Studies of natural rewards have also shown that NAc neurons exhibit changes in firing during reward consumption (Schultz 2000; Nicola et al. 2004). In contrast to the phasic activity time-locked to reward-predictive cues and operant behavior, the majority of the changes in firing during reward consumption are inhibitory. Though cocaine self-administration is not associated with specific consummatory behaviors, a high percentage (50%) of neurons show an inhibition in firing during the min(s) following each cocaine infusion. The nature of these decreases is not yet understood; however, it is possible that the change in firing is analogous to the inhibitory response exhibited by NAc neurons during consumption of natural rewards.

2.5.3. Long-Duration Firing Patterns Time-Locked to the Cocaine-Reinforced Lever Press: the Progressive Reversal Firing Pattern

2.5.3.1. Description of the Firing Pattern

NAc neurons exhibit another category of firing pattern time-locked to the cocaine-reinforced lever press (Peoples et al. 1994, 1998b; Peoples and West 1996). At an intermediate dose of cocaine (0.75 mg/kg/inf), approximately 65% of all neurons show a change in firing that occurs slowly across the interval between successive self-infusions. For most neurons (i.e., 50% of all recorded neurons), firing rate decreases during the first minute postpress and then progressively increases until the time of the next lever press (i.e., decrease + progressive reversal) (Figure 2.10). For the remaining neurons, firing rate increases postpress and progressively decreases until the time of the next press (i.e., increase + progressive reversal) (Figure 2.10).

FIGURE 2.10. The progressive reversal firing patterns time-locked to the cocaine-reinforced lever press during an FRI cocaine self-administration session.

FIGURE 2.10

The progressive reversal firing patterns time-locked to the cocaine-reinforced lever press during an FRI cocaine self-administration session. The figure shows single-neuron examples of a decrease + progressive reversal firing pattern (top) and an increase (more...)

2.5.3.2. Determinants of the Firing Patterns

Investigation of the progressive reversal firing pattern exemplifies some of the methods that can be used to differentiate between nonpharmacological and pharmacological determinants of firing patterns observed during periods of drug exposure. In particular, the research demonstrates use of the following: (1) time course comparisons of firing patterns and pharmacological events, (2) time course comparisons of firing patterns and nonpharmacological events, and (3) the behavioral clamp method. The research also highlights the types of nonpharmacological variables that potentially contribute to NAc firing patterns during drug self-administration.

The decrease + progressive reversal firing pattern parallels changes in locomotion and stereotypy that occur over the course of the interinfusion interval (i.e., between successive self-infusions) (Figure 2.11). Peoples et al. (1998b) tested whether the decrease + progressive reversal firing pattern reflected changes in the frequency of locomotion. To test this interpretation we determined whether neurons that show the decrease + progressive reversal firing pattern also showed phasic changes in firing time-locked to specific locomotion events. Moreover, we used the clamping procedure to test whether the firing pattern depended on the variations in locomotion over the course of the interinfusion interval.

FIGURE 2.11. Average percentage of time spent in locomotion and stereotypy exhibited by all subjects during an FR1 cocaine self-administration session.

FIGURE 2.11

Average percentage of time spent in locomotion and stereotypy exhibited by all subjects during an FR1 cocaine self-administration session. Average time spent (per 30-sec bin) is plotted as a function of min pre- and postpress. Time 0 = time of reinforced (more...)

We carried out these tests with the aid of video analysis. During each recording session, behavior was videotaped. Each video frame (33-msec) was sequentially time-stamped by a computer coupled with a video frame counter. Frames were time-stamped according to the same computer clock that time-stamped each neural discharge. After the recording sessions, in offline frame-by-frame analysis, time stamps associated with the onsets and offsets of particular behaviors (33-msec temporal resolution) were compiled and used in subsequent histogram analysis of neural firing in relation to the specific behaviors.

Comparisons of average firing during locomotion showed that half of the decrease + progressive reversal neurons showed a significant increase in average firing during locomotion relative to the seconds immediately preceding the onset of the behavior. For almost all of these neurons, this phasic response was selective for either locomotion toward the lever or locomotion away from the lever. The absence of phasic firing for half the neurons and the directional specificity of the phasic locomotion responses for the other neurons showed that the decrease + progressive reversal firing patterns are not related to locomotion per se. The clamping control provided additional evidence against the motor interpretation. Specifically, when histograms displaying firing during the minute before and after cocaine infusion were reconstructed with all periods of locomotion excluded from the calculation of firing rate, none of the decrease + progressive reversal patterns were lost (Figure 2.12). Thus, none of the decrease + progressive reversal firing patterns could be attributed to variations in the frequency of locomotion across the minutes of the interinfusion interval.

FIGURE 2.12. The change + progressive reversal firing patterns persisted when all locomotion was excluded from calculations of firing rate.

FIGURE 2.12

The change + progressive reversal firing patterns persisted when all locomotion was excluded from calculations of firing rate. Each histogram shows phasic firing time-locked to the cocaine-reinforced lever press. Histograms shown in the left column correspond (more...)

Though the firing patterns are not related to movement per se, the above findings show that some of the decrease + progressive reversal firing patterns may be related to the directional, spatial characteristics of locomotion. Alternatively, the patterns, like lever-press firing patterns, might be related to reward expectations. This interpretation is consistent with evidence that mesolimbic DA and the NAc do not directly mediate the execution of movements; instead they have a psychomotor or motivational function that facilitates incentive-related approach and preparatory behaviors (Iversen and Koob 1977; Robbins and Everitt 1982; Beninger 1983; Kelley and Stinus 1985; Wise and Bozarth 1987; Cador et al. 1989, 1991; Everitt et al. 1989; Apicella et al. 1991; Di Chiara et al. 1992; Salomone 1992; Schultz et al. 1993).

In considering possible reward-related determinants of the decrease + progressive reversal firing pattern, it is relevant to note that the firing pattern mirrors changes in drug and DA levels during the cocaine self-administration session (Wise et al. 1995; Peoples and West 1996). The changes in drug level that occur between successive self-infusions are expected to engender interoceptive cues (Colpaert 1987; Overton 1987; Stewart and deWit 1987), including those that develop progressively over the course of the interval, in conjunction with drug metabolism. These interoceptive cues could include or become associated with changes in the motivational state of the animal that occur with changes in drug level. It is possible that the progressive increase in firing reflects excitatory afferent input related to the cues and motivational state of the animal and that the rapid decrease in firing that occurs postpress reflects the cessation of that afferent input. An alternative, but not mutually exclusive, interpretation of the firing pattern is that it reflects more direct acute actions of cocaine in the NAc. The relative contribution of pharmacological and nonpharmacological variables to the decrease + progressive reversal firing patterns remains to be further delineated. However, there is evidence that the patterns cannot be wholly pharmacological. For example, for some neurons the postpress decrease begins before the onset of the cocaine infusion (Peoples and West 1996; Peoples et al. 1999b).

Regardless of the relative contribution of pharmacological and nonpharmacological factors to the firing pattern, we hypothesize that the pattern contributes to regulation of the self-administration behavior. This hypothesis is based on the following observations. As already noted, the pattern closely mirrors changes in drug and accumbal DA levels. Previous research has shown that successive self-infusions occur predictably, at time points when drug and accumbal DA level falls to a particular level. Single trial analysis of the firing rates of decrease + progressive reversal neurons shows that the timing of self-infusions is also associated with the neurons' achieving a fairly reliable apical firing rate. Additional analyses showed that the duration of the progressive increase in firing is positively and significantly correlated with the interinfusion interval: The longer the interinfusion interval, the longer the duration of the progressive increase in firing. The decrease + progressive reversal firing pattern also parallels changes in the propensity of animals to approach the response lever. Based on these observations and the findings of the locomotion and behavioral clamp analyses, the decrease + progressive reversal firing pattern likely influences the propensity of animals to initiate drug-directed behavior.

2.5.4. Session-Long Changes in Average Firing Rate

2.5.4.1. Description of the Firing Patterns

Extracellular recordings during intravenous FR1 cocaine self-administration (0.75 mg/kg/inf) sessions have shown that most (≈90%) NAc neurons exhibit a change in average firing rate during the self-administration session relative to the pre- and post-session baseline periods (referred to herein as session-long increases and decreases, Figure 2.13) (Chang et al. 1998; Peoples et al. 1998a, 2004, 2007b). The session-long changes in firing are exhibited by neurons that exhibit a phasic lever-press firing pattern, a progressive reversal firing pattern, or both. But, the patterns are also exhibited by neurons that show no other change in firing (Peoples et al. 1998b). At the dose of 0.75 mg/kg/inf, the majority of the session-long changes in firing (i.e., 60% percent of all recorded neurons) are decreases.

FIGURE 2.13. Session-long changes in firing during FR1 cocaine self-administration sessions.

FIGURE 2.13

Session-long changes in firing during FR1 cocaine self-administration sessions. The figure shows single-neuron examples of a session-long increase (left) and a session-long decrease (right) in firing. In each histogram, average firing (Hz per 30-sec bin) (more...)

2.5.4.2. Determinants of the Firing Pattern

The session-long change in firing pattern is defined by a sustained change in average firing during the self-administration session relative to the drug-free baseline period. It thus involves a difference in firing between a drug-free period and the entire period of drug exposure and potentially reflects an acute drug effect. Though the firing pattern has a time course consistent with a potential pharmacological origin, it is also possible that the firing patterns reflect NAc encoding of nonpharmacological events. The self-administration session and the baseline periods differ in a number of ways. During the baseline period animals generally remain at rest, locomoting only occasionally. However, during the self-administration session animals engage in operant behavior and exhibit increased locomotion and drug-induced stereotypic behaviors. The session-long changes therefore potentially reflect pharmacological actions of cocaine, afferent input associated with the behaviors that are unique to the self-administration session, or both.

A number of studies have been conducted to differentiate among the various interpretations of the session-long change in firing. The investigations provide examples of numerous methods that can be employed to differentiate pharmacological and nonpharmacological firing patterns. These include the following: (1) time course comparisons between the firing patterns and pharmacological events, (2) time course comparisons between the firing patterns and nonpharmacological events, (3) comparisons of firing patterns during drug-directed behavior and similar natural reward-directed behavior, (4) behavioral clamp procedures, and (5) comparisons with natural reward. The description of the firing patterns is presented in two sections of this chapter. Here we limit the discussion to research that demonstrates a nonpharmacological component to the firing patterns. Research related to pharmacological determinants of the firing patterns is overviewed in Section 2.7.

Two lines of evidence are consistent with the hypothesis that at least some portion of the session-long changes in firing is nonpharmacological. First, the onset of some of the changes in firing precedes the delivery of the first drug infusion and occurs in association with the cues that signal the onset of drug availability and the onset of the first operant response (Peoples et al. 2004). Second, NAc neurons exhibit session-long changes in firing during FR1 sucrose-reinforced operant behavior (Kravitz et al. 2006; Kravitz and Peoples 2008). These observations are consistent with the hypothesis that at least some of the session-long changes in firing reflect nonpharmacological processes.

In considering possible nonpharmacological determinants of the firing patterns, we first tested the hypothesis that the patterns reflect an effect of phasic firing time-locked to the operant behavior on average firing. Several lines of evidence led to the rejection of this hypothesis. First, the majority of session-long changes in firing do not exhibit a phasic change in firing time-locked to the lever-press operant. Second, in 75% of the cases in which phasic and session-long changes co-occur, the changes are directionally opposed (Peoples et al. 2004; unpublished observations). Third, session-long changes in firing are apparent regardless of whether firing rate calculations are limited to either periods in which the animal is engaged in drug-directed behavior, or periods in which other behaviors such as stereotypy occur (Peoples et al. 2004 and Figure 2.14). These observations are consistent with the interpretation that session-long changes in firing do not reflect an effect of phasic firing time-locked to operant behavior, stereotypy, or locomotion.

FIGURE 2.14. Average firing rates of session-increase (top) and session-decrease (bottom) neurons during specific 30-sec periods before and after each lever press of a single cocaine self-administration session.

FIGURE 2.14

Average firing rates of session-increase (top) and session-decrease (bottom) neurons during specific 30-sec periods before and after each lever press of a single cocaine self-administration session. Firing is shown during the -120 to -90 sec before each (more...)

2.5.5. Firing Patterns During Cocaine Taking: Section Summary, Novel Findings, and Research Opportunities

One goal of experiments that employ the chronic microwire recording method is to identify acute pharmacological effects on NAc neural activity, which might contribute to cocaine addiction. In behaving animals, a change in single neuron activity that occurs with a change in drug exposure can reflect the following: (1) pharmacological actions on the recorded neurons or primary afferent inputs, or (2) neural responses to differences in behavior, stimulus events, or other factors between the drug-exposed and drug-free condition. For behaving animal recordings to be useful in identifying pharmacological mechanisms that contribute to addiction, it is necessary for those studies to differentiate the relative contribution of the pharmacological and nonpharmacological variables to changes in firing patterns associated with drug exposure. A number of strategies can be employed to make this differentiation. The research reviewed in this section exemplifies a number of them. Additional approaches are described in later sections of the chapter. The research also identified two firing patterns that potentially reflect acute actions of self-administered cocaine. Subsequent sections describe additional chronic microwire recording studies of the pharmacological effects of cocaine on NAc neurons.

An additional potential application of the chronic microwire recording method is to investigate drug-induced changes in information processing that may underlie the abnormal motivated behavior that characterizes addiction. It is possible that the development of the behaviors is mediated in part by acute drug-induced alterations in information processing, which influence normal adaptive processes associated with learning and memory. If this were the case one might expect to observe clear differences in event-related NAc neural responses during operant behavior that occurs in the presence versus absence of cocaine.

Available evidence shows that there are both similarities and differences between task-related neural activity during cocaine- and natural reward-directed instrumental behavior. Some of the similarities include the predominance of excitatory changes in firing time-locked to predictive cues and instrumental behavior, and evidence that the firing patters are related to reward-related expectations. Thus far, the most prominent differences include the greater prevalence of session-long decreases in firing during drug-reinforced operant sessions, the progressive reversal firing pattern, and the dissociation between neurons that show task-related firing during behavior directed toward cocaine and natural rewards. The functional significance of these differences is not fully understood. Further evaluation of the question has implications for understanding the role of acute drug actions in the development of drug addiction. Strong overlap would suggest that acute drug-induced changes in information processing play a limited role; whereas significant differences suggest that those drug actions might play an important role in the development of addiction.

2.6. FIRING PATTERNS DURING COCAINE SEEKING

2.6.1. Overview of Section

In addition to characterizing firing patterns during FR1 cocaine self-administration sessions, researchers have begun to document and investigate firing patterns during periods in which animals engage in drug-directed behavior under drug-free conditions. There are a number of reasons for conducting these types of studies. First, neuropharmacological studies show that the mechanisms that mediate drug-directed behavior under drug-free and drug-exposed conditions are not identical (Whitelaw et al. 1996; Arroyo et al. 1998; Markou et al. 1999; Grimm and See 2000; Olmstead et al. 2000, 2001; Everitt et al. 2001; Cardinal et al. 2002; Dickinson et al. 2002). Second, drug-directed behavior under drug-free conditions (referred to as drug-seeking) is thought to better model conditions associated with relapse in humans, which is a primary therapeutic target (O'Brien 2005). Third, the research also represents an additional approach to differentiating firing patterns that are mediated by acute drug actions from those that are nonpharmacological. NAc firing patterns associated with drug-directed behavior in drug-free animals have been characterized in three basic types of experimental situations. The experiments and findings are described in this section with the intent of providing further overview of methods but also to highlight novel findings and a unique research opportunity afforded by the chronic-behaving animal recording procedure.

2.6.2. Experiments

2.6.2.1. NAc Firing at the Onset of an FR1 Cocaine Self-Administration Session

In one study we characterized NAc responses during cue-evoked drug-directed behavior in drug-free animals. Recordings were made in animals with a two-week history of daily cocaine (0.75 mg/kg/inf) self-administration (Peoples and Cavanaugh 2003; Peoples et al. 2004). The recording session consisted of a 20-min presession baseline period, a cocaine self-administration session, and a 40-min postsession baseline period. During the presession baseline and postsession recovery period, animals did not have access to the operant response lever, and the drug was not available. The onset of the cocaine self-administration was signaled by presentation of a compound tone (7.5 sec) + light (30 sec) cue. The cue was the same as that typically associated with presentation of the primary cocaine reinforcer. Animals initiated cocaine self-administration shortly after the onset of the cue, but not within less than 1 min.

Average firing during the 30-sec cue that signaled the onset of the self-administration session (referred to as the discriminative stimulus, SD) was compared with firing during the 30-sec preceding the cue. Changes in firing during the 30-sec cue presentation relative to the 30-sec precue period were referred to as SD changes in firing. We additionally compared average firing during the 30 sec that preceded the first cocaine-reinforced operant response and the 30-sec pre-SD. Changes in firing during this period were referred to as pre-first-press responses.

The findings of this study showed that about 20% of all recorded NAc neurons respond during presentation of the cue. About 60% of the cue-responsive neurons also show a first-press response. An additional 10% of all recorded neurons show a first-press firing pattern without showing an SD response. Almost all (>85%) of the SD and pre-first-press responses are excitatory (Figures 2.152.16). In many cases, SD responses are sustained during the minutes that elapse between the onset of the SD and the occurrence of the first press. The sustained nature of the SD firing pattern is consistent with the interpretation that for a majority of neurons; the SD response is not related to the physical properties of the cue. The timing of the firing patterns is also dissociable from that of specific behaviors during the cue and also during the 30-sec pre-first press (e.g., Figure 2.15). The total percent of neurons that show either an SD or first-press response is consistent with the prevalence of neurons that show lever-press firing patterns during the cocaine self-administration session. Moreover, the majority of neurons that show either an SD or pre-first-press firing pattern also show a phasic increase in firing time-locked to the cocaine-reinforced lever press.

FIGURE 2.15. Excitatory changes in firing time-locked to presentation of cues that signaled the onset of an FR1 cocaine self-administration session.

FIGURE 2.15

Excitatory changes in firing time-locked to presentation of cues that signaled the onset of an FR1 cocaine self-administration session. The change in firing time-locked to the onset of a cue that signaled the start of a cocaine self-administration session (more...)

FIGURE 2.16. Excitatory changes in firing associated with the onset of drug-directed behavior during an FR1 cocaine self-administration session.

FIGURE 2.16

Excitatory changes in firing associated with the onset of drug-directed behavior during an FR1 cocaine self-administration session. The figure shows two single-neuron examples of changes in firing time-locked to the 1st cocaine-reinforced press (i.e., (more...)

2.6.2.2. NAc Firing Patterns during Cue-Evoked and Nonreinforced Drug-Directed Behavior

In another study (Ghitza et al. 2003), NAc recordings were conducted during periods in which animals were first presented with a drug-associated cue and then initiated a period of unreinforced drug-directed operant behavior (i.e., a period of extinction). In the study, Ghitza et al. (2003) trained rats on a tone discrimination paradigm. Specifically, during the self-administration sessions, individual lever presses that occurred during tone presentation produced an intravenous infusion of cocaine (0.35 mg/kg/inf) and terminated the tone. Presses made in the absence of the tone were nonreinforced. Once animals showed stable operant responding, animals were exposed to a 3–4 week period of abstinence. Animals were then returned to the operant chamber and exposed to a single presentation of the cocaine-predictive tone. NAc single-unit activity was recorded during re-exposure to the tone under extinction conditions (i.e., operant behavior was not reinforced and animals remained drug free).

Of the neurons recorded during extinction, 37 neurons (53%) exhibited at least a twofold, tone-evoked change in firing within 150 msec after tone onset, relative to baseline firing during the 150-msec time period preceding tone onset. Of the NAc neurons exhibiting tone-evoked activity, 76% were excited and 24% were inhibited by the tone. Although the tone-evoked change in firing of these neurons commenced before the onset of drug-directed behavior, the change in firing persisted after the onset of the behavior for 86% of the neurons. Control analyses showed that the changes in firing were not attributable to motor behavior and did not reflect a primary sensory response.

2.6.2.3. Drug-Directed Behavior Maintained by Presentation of a Conditioned Reinforcer

Studies of NAc dopamine (DA) show that the accumbal response to the presentation of unexpected, experimenter-delivered cue presentations is not the same as that associated with presentation of expected, response-conditioned presentations (Ito et al. 2000, 2002). The behavior of humans is thought to be partially directed and maintained by positive conditioned predictors and conditioned reinforcers (O'Brien et al. 1990; Childress et al. 1993; Tiffany 1999; Garavan et al. 2000). Based on these observations, it may be important to characterize NAc neural activity in relation to drug-directed behavior maintained by conditioned reinforcers. We have begun to approach this issue using a multi-phase seeking-taking task.

The multi-phase task is conducted as follows. During the first phase of the two-phase task, animals engage in operant behavior reinforced on an FR10(FR10:S) schedule of reinforcement (one to five trials). Under this second-order schedule, animals earn 10 conditioned reinforcer (i.e., S) presentations on an FR10 schedule of reinforcement. On this schedule of conditioned reinforcement, every 10th press was reinforced by presentation of a tone + light cue (i.e., the conditioned reinforcer). The 10th presentation of the conditioned reinforcer is paired with a cocaine infusion. Thus, each cocaine infusion is preceded by the completion of 100 operant responses. Animals earn five reinforcers on this schedule of reinforcement during the first, “seeking” phase of the session. During the second phase of the session, an FR1 TO 60 sec schedule was in effect. This schedule establishes a substantive period of drug-free drug-directed behavior (i.e., during the first trial of the session). This procedure is meant to mimic periods of drug-free drug-directed behavior that might precede a period of drug taking in humans.

In one experiment, we trained four rats to self-administer cocaine (0.75 mg/kg/ inf) using the multiphase task. Analysis of NAc firing showed that a subset of neurons exhibited a sustained change in average firing during the first, drug-free trial of the self-administration session. For some neurons (about 30% of recorded neurons) the change was an increase, but an equal number of neurons showed a decrease in average firing (Figure 2.17). In addition to the sustained changes in firing, we observed that a subset of neurons exhibited phasic changes in firing time-locked to delivery of the conditioned reinforcer (Figure 2.18). These changes in firing were exclusively excitatory. The neurons that showed a phasic response time-locked to the conditioned reinforcer also showed an average excitatory response time-locked to the cocaine-reinforced operant during the subsequent “taking” phase of the experiment. Though no phasic decreases in firing occurred time-locked to the conditioned reinforcer, there were a small number of neurons that showed such a response time-locked to the cocaine-reinforced operant during the taking phase. In a control study, we trained four additional animals to self-administer sucrose during the two-phase seeking-taking task. Firing patterns during the first, seeking phase of the session were comparable for the sucrose- and cocaine-trained animals (Figures 2.192.20).

FIGURE 2.17. Average firing of NAc neurons during cocaine- and sucrose-directed operant behavior reinforced by presentation of a conditioned reinforcer.

FIGURE 2.17

Average firing of NAc neurons during cocaine- and sucrose-directed operant behavior reinforced by presentation of a conditioned reinforcer. The figure shows average firing rates for three groups of neurons recorded during multiphase seeking-taking operant (more...)

FIGURE 2.18. Group-mean lever-press firing patterns during periods of conditioned reinforcement and primary cocaine reinforcement.

FIGURE 2.18

Group-mean lever-press firing patterns during periods of conditioned reinforcement and primary cocaine reinforcement. Average firing is shown for two groups of neurons recorded during a multiphase seeking-taking cocaine self-administration session. The (more...)

FIGURE 2.19. Group-mean phasic firing patterns during a multiphase surcrose seeking-taking task.

FIGURE 2.19

Group-mean phasic firing patterns during a multiphase surcrose seeking-taking task. Average firing is shown for two groups of neurons recorded during a multiphase seeking-taking cocaine self-administration session. The groups include the following: (1) (more...)

FIGURE 2.20. Average firing of session-long decrease neurons decreases progressively across the initial cocaine self-infusions of the session.

FIGURE 2.20

Average firing of session-long decrease neurons decreases progressively across the initial cocaine self-infusions of the session. Firing of specific accumbal neurons decreases as cumulative drug intake increases during the loading phase of cocaine self-administrations. (more...)

2.6.3. Firing Patterns During Cocaine Seeking: Summary, Novel Findings, and Research Opportunities

Drug-directed behavior that occurs in the absence and presence of drug is referred to as drug seeking and drug taking, respectively. Various lines of work indicate that the mechanisms that mediate drug seeking and taking are not the same. Drug seeking is generally viewed as more relevant to relapse, whereas drug taking is more relevant to pharmacological events that mediate the development of addiction. The dissociation is viewed as important to successful investigation of the causes and treatment of addiction.

In this section we reviewed three experiments that characterized NAc activity during cocaine seeking. Two of the three studies characterized NAc responses to an initial unpredicted presentation of a drug-associated cue and the subsequent interval leading up to the first operant (Peoples et al. 2007a). One of these studies additionally characterized operant responding during a period of extinction. Despite the multiple parametric differences among the three studies, the findings were quite consistent in showing that the NAc responses associated with the presentation of drug-associated cues and the onset of drug-directed behavior are predominantly excitatory and of a sustained nature. A different profile of NAc firing was observed in animals engaged in drug seeking maintained by conditioned reinforcement. In particular, there was a much higher prevalence of sustained decreases in average firing during the period of drug-free drug-directed behavior. The greater mix of sustained increases and decreases, as well as the stability of phasic firing patterns time-locked to operant behavior during the first drug-free trial of conditioned reinforcement and the primary reinforcement (FR1) phase, suggests that patterns of NAc neural activity share more similarities between drug-free and drug-exposed periods when drug seeking is maintained by conditioned reinforcement than when the behavior occurs during extinction conditions.

The studies of drug seeking have potential implications for investigations that aim to identify causes of relapse and to screen for potential preventative medicines. Most animal studies of neurobiological determinants and potential medications for relapse employ extinction-reinstatement models. Some researchers have observed that the presence versus absence of a history of extinction can alter the neuronal circuits that are involved in reinstatement of drug-directed behavior (Fuchs et al. 2006). The present findings show that NAc neural activity is different during extinction versus conditioned, reinforced drug-directed behavior. Based on these findings, it would seem important to expand studies of relapse to include procedures that evaluate drug seeking under nonextinction conditions.

2.7. ACUTE PHARMACOLOGICAL EFFECTS OF SELF-ADMINISTERED COCAINE

2.7.1. Background and Overview

Electrophysiological recordings in slice and anesthetized animal preparations show that experimenter-delivered cocaine inhibits spontaneous and glutamate-evoked firing of NAc neurons (Qiao et al. 1990; Uchimura and North 1990; White et al. 1993; Nicola et al. 1996). This inhibitory effect is mediated primarily by cocaine-induced elevations of accumbal DA and associated increases in activation of DA receptors (for review, see Nicola et al. 2000; also see White et al. 1993, 1998; Hu and White 1994; Henry and White 1995). Ongoing efforts in our laboratory are designed to test whether the acute pharmacological actions of cocaine are also inhibitory.

In testing this possibility we have focused on the session-long changes in firing, which reflect changes in firing during drug exposure relative to pre- and post-session drug-free periods. As already described (2.5.4.2) both the time course of some of the session-long changes and the presence of similar types of changes during sucrose self-administration sessions indicate that normal afferent input associated with reward-directed behavior contribute to the session-long changes in firing. However, the session-long inhibitions tend to be more prevalent during drug self-administrations as compared to natural reward self-administration sessions. It has thus been hypothesized that the session-long changes in firing during cocaine self-administration sessions additionally reflect pharmacological actions of cocaine.' Published and preliminary findings of these efforts are described in this section. The reviewed research demonstrates a number of methods that can be used in identifying pharmacological effects of self-administered drug, including (1) time course comparisons of firing patterns and changes in drug level; (2) dose-response studies; and (3) comparisons to natural reward-directed behavior. The research also demonstrates the importance of extending electrophysiological studies of pharmacological cocaine effects beyond slice and anesthetized animal procedures to include recordings in behaving animals.

2.7.2. Session-Long Decreases but not Session-Long Increases in Firing Covary With Drug Level

We have made two observations consistent with the interpretation that session-long decreases but not session-long increases are dose-dependent. First, at the beginning of a self-administration session, drug level rises progressively across the first few infusions and then attains a level that is maintained for the duration of the session. Consistent with a possible relationship between session-long inhibition and drug level, onset of the firing pattern, on average, tends to be progressive: firing decreases across the first several infusions of cocaine and then attains a minimum that is maintained for the rest of the session (Figures 2.14, 2.20) (Peoples et al. 1998b, 1999b, 2007b).

Second, a preliminary dose-response study showed that NAc inhibition but not excitation increases with dose of self-administered cocaine. We trained two animals to self-administer cocaine on an FR1 schedule of cocaine reinforcement (0.75 mg/ kg/inf) and then exposed each animal to two test sessions. In the first session, a within-session dose-response curve was determined by incrementing the dose of self-administered cocaine according to the following schedule: 0.187, 0.375, 0.75, and 1.5 mg/kg/inf. During the second test session animals self-administered the same doses in a descending order.

A total of 12 NAc neurons were recorded during the two sessions. Animals self-administered 15 infusions of each cocaine dose. To characterize the effect of dose on firing, we calculated firing rate of each neuron during the period that lapsed between the 6th and the 15th infusions and compared these average firing rates to average firing during the presession baseline period (Mann-Whitney tests, α = 0.05). Between-dose comparisons showed that both the prevalence and magnitude of inhibitory responses were positively related to dose of self-administered cocaine. There was no effect of incrementing the cocaine dose on the excitatory changes in average firing (Figure 2.21). The positive relationship between drug level and the prevalence and magnitude of inhibitory changes in average firing is consistent with a potential contribution of drug actions to the firing patterns. However, changes in drug level are associated with changes in reward magnitude (Spear and Katz 1991; Thomsen and Caine 2006). It is thus necessary to consider the possibility that dose-dependent changes in the session-long decreases reflect NAc encoding of changes in reward rather than a neurophysiological action of cocaine.

FIGURE 2.21. Changes in average firing of NAc neurons during a within-session dose-response curve.

FIGURE 2.21

Changes in average firing of NAc neurons during a within-session dose-response curve. Animals trained to self-administer cocaine (FR1 TO 60 sec; 0.75 mg/kg) were exposed to one of three doses (0.75, 1.5, or 3.0 mg/kg/inf). On the left, prevalence of session-decrease (more...)

If the behavioral interpretation were correct, one would expect that increments in reward magnitude, in the absence of drug, would have an effect on NAc firing that was similar to that of incrementing drug dose. We tested this prediction in one experiment. Three groups of animals were trained to self-administer 4%, 32%, or 64% sucrose (0.2 ml reinforcer) on an FR1 TO 60-sec schedule. Recordings were conducted on the 10th day of training. The number of neurons recorded at each of the three sucrose concentrations equaled 32, 51, and 38, respectively. Between-group comparisons showed that increasing sucrose concentration did not increase the prevalence or magnitude of session-long decreases that occurred during the sucrose self-administration session. However, there was some evidence that the prevalence of session-long increases were more prevalent at the 32% and 64% concentrations relative to the 4% concentration (Figure 2.22). The findings support the interpretation that the session-long decreases reflect something other than NAc tracking of reward magnitude. The relation between session-long increases and reward magnitude is less clear but may be positive.

FIGURE 2.22. Tonic firing patterns during sucrose self-administration.

FIGURE 2.22

Tonic firing patterns during sucrose self-administration. Prevalence (left) and magnitude (right) of tonic firing patterns during FR1 self-administration of varying concentrations of sucrose. Data recorded during FR1 60-sec sessions.

Comparable findings to this dose-response study were found in a study of second-order self-administration. Four animals were trained on the FR10(FR10:S) procedure that we have already described. Animals earned a total of five cocaine reinforcers on this second-order schedule before initiating the second FR1 phase of the experiment. During the first second-order trial, the percent of neurons that showed tonic increases and decreases were similar and less than 30%. The percent of neurons that showed a significant decrease in firing relative to presession baseline increased across the subsequent four trials and increased further during the second, FR1 TO 60-sec phase of the session. The progressive increase in inhibition occurred with progressive increases in drug level (Figure 2.23, top). The converse was true for increases in firing. Consistent with these changes, the average firing rate of neurons decreased across the five second-order trials (Figure 2.23, middle). Control analyses showed comparable results when between-trial comparisons of firing rate were limited to periods in which animals engaged in particular behaviors (Figure 2.24). The between-trial changes in firing thus did not depend on between-trial differences in prevalence of particular behaviors. The prevalence and magnitude of tonic increases and decreases in firing during the second phase of the session were comparable to those observed in previous experiments that used the FR1 TO 60-sec procedure. The study shows that inhibition of NAc neurons during drug-directed behavior is greater during periods of acute cocaine exposure than during drug-free periods. Moreover, inhibition of NAc neurons during drug-directed behavior increases as the level of cocaine rises (i.e., is dose dependent). These findings are consistent with those that one expects to observe if the acute pharmacological effect of cocaine on NAc neurons is inhibition.

FIGURE 2.23. Average firing rate across 5 FR10(FR10:S) trials.

FIGURE 2.23

Average firing rate across 5 FR10(FR10:S) trials. Top and middle: Drug level and average firing rate of three groups of NAC neurons during 5 FR10:(FR10:S) cocaine-reinforced trials. The three groups correspond to neurons that showed either an increase, (more...)

FIGURE 2.24. Firing during different periods of FR10(FR10:S) trials.

FIGURE 2.24

Firing during different periods of FR10(FR10:S) trials. Firing rates are shown for three groups of neurons: those that showed an increase (top), decrease (middle), or no change (bottom) in firing during the first FR10(FR10:S) trial. Firing rates were (more...)

A similar experiment was conducted in sucrose-trained animals. Four animals were trained to self-administer sucrose in the multiphase sessions. As for Aim 2 Exp 2, a second-order schedule FR10(FR10:S) was in effect during the first phase of the session and a first-order FR1 TO 60-sec schedule was in effect during the second phase of the session. During the first second-order trial, the prevalence of tonic increases and decreases were similar and less than 30%. Between the first second-order trial and the subsequent second-order trials, as well as the FR1 TO 60-sec phase, tonic changes in firing, and hence average firing rates, were highly stable (Figure 2.23 bottom). The findings are consistent with the interpretation that session-long NAc inhibition during sucrose-directed behavior is not affected by the presence versus absence of the primary reward. Inhibition of NAc firing during cocaine self-administration sessions is thus difficult to attribute to a normal nonpharmacological response of NAc neurons to reward exposure.

2.7.3. Evidence that the Inhibitory Effects of Self-Administered Cocaine are Activity Dependent

DA is hypothesized to have activity-dependent effects on the firing of NAc neurons. Specifically, it has been hypothesized that dopamine facilitates the transmission of strong excitatory input through the NAc, while simultaneously suppressing weak inputs (Rolls et al. 1984; Morgenson and Yim 1991; Pennartz et al. 1994; Pierce and Rebec 1995; Kiyatkin and Rebec 1996; Levine et al. 1996; O'Donnell and Grace 1996; Hernandez-Lopez et al. 1997). Given the importance of DA in cocaine effects on NAc neurons in anesthetized animal and slice studies, we have hypothesized that inhibition of NAc neurons during cocaine self-administration sessions might be activity dependent. Specifically, during cocaine self-administration sessions, neurons that receive excitatory afferent input associated with task-related events might be less susceptible to the inhibitory effects of self-administered cocaine. For example, neurons that are phasically activated during cocaine self-administration sessions might, as a group, show less inhibition than neurons that show no phasic excitatory activity. Evidence consistent with this expectation has been observed in a number of studies.

In one of the studies (Peoples et al. 2007b), rats were trained in a small number of short sessions to self-administer cocaine on an FR1 schedule of reinforcement (0.75 mg/kg/inf). Animals were then exposed to a 30-day regimen of daily long-access (LgA; 6 hr/day) cocaine self-administration. Recordings were conducted on the second, third, and 30th day of cocaine self-administration. Recorded neurons were sorted into three groups: (1) all those that showed a session-long increase in firing (session-activated neurons, e.g., Figure 2.13A); (2) those that showed a phasic excitatory response to a task event, including the cue that signals the onset of the session and the first and subsequent cocaine-reinforced lever presses (i.e., referred to as event-but-not-session-activated neurons; e.g., Figures 2.6, 2.15, 2.16); and (3) those that showed neither a session-long increase nor a response to any task event (i.e., task-nonactivated neurons).

Within each recording session, the three groups of neurons maintained different average firing rates (Figure 2.25, top versus middle versus bottom rows). Specifically, average firing rate of the session-activated neurons was significantly greater than average firing of the event-but-not-session-activated neurons, and the average firing rate of each of those neuron groups was greater than that of the task-nonactivated neurons. The different rates of average firing appeared to reflect a differential sensitivity of the neuron groups to acute inhibitory effects of cocaine. For example, the event-but-not-session-activated neurons, as a group, maintained average firing during the cocaine self-administration session relative to a presession baseline period. This was true even if phasic firing was excluded from calculations of firing rate. In contrast, the task-nonactivated neurons showed a significant decrease in average firing during the cocaine self-administration session relative to the presession baseline period. Similar findings were observed in an earlier study (Peoples et al. 2004). The findings are consistent with the interpretation that the acute inhibitory actions of self-administered cocaine are activity dependent, though additional studies are required to determine whether the activity-correlated changes in firing reflect activity-dependent, DA-mediated actions on the recorded neurons.

FIGURE 2.25. Average firing of groups of neurons during an early day (Day 2-3) and a late day (Day 30) of the 30-day regimen of long-access (6-hr) cocaine self-administration.

FIGURE 2.25

Average firing of groups of neurons during an early day (Day 2-3) and a late day (Day 30) of the 30-day regimen of long-access (6-hr) cocaine self-administration. Firing is shown during three periods for each session: (1) the last 5 min of the presession (more...)

2.7.4. Acute Pharmacological Effects of Self-Administered Cocaine: Summary, Novel Findings, and Research Opportunities

Electrophysiological recordings generally show that experimenter-delivered cocaine inhibits spontaneous and glutamate-evoked firing of NAc neurons in both the slice and anesthetized animal preparations (Qiao et al. 1990; Uchimura and North 1990; White et al. 1993; Kiyatkin and Rebec 1996; Nicola et al. 1996). The findings of our behaving animal recordings are consistent with the interpretation that the primary pharmacological acute action of self-administered cocaine is also inhibition. However, the behaving animal studies show that not all neurons are inhibited during cocaine self-administration. In fact, a substantive portion of neurons show session-long increases in firing, phasic increases in firing, or both (Peoples et al. 1998b, 2007b). Moreover, the magnitude of inhibition induced by self-administered cocaine is reduced in neurons that receive excitatory afferent input (i.e., are phasically activated) relative to neurons that do not receive such afferent input during drug exposure. The substantial portion of noninhibited neurons and the possible activity-dependent nature of cocaine-induced inhibition are not predicted by the slice and anesthetized animal studies of acute cocaine effects on NAc neurons. The findings are likely to have important implications for understanding acute effects of cocaine that are relevant to cocaine addiction.

2.8. CHRONIC DRUG-INDUCED NEUROADAPTATIONS

2.8.1. Slice and Anesthetized Animal Recording Studies

Multiple electrophysiological studies have shown that a history of repeated cocaine exposure induces hypoactivity. In anesthetized animals, the responsiveness of NAc neurons to excitatory stimulation is reduced in animals with a history of experimenter-delivered cocaine (White 1992; White et al. 1995a). Similarly, in NAc slices, excitatory currents and whole cell excitability are decreased in rats with a history of cocaine exposure (Zhang et al. 1998). Moreover, repeated injections of cocaine can lead to a decrease in the ratio of AMPA to NMDA (i.e., AMPA:NMDA) currents, a mechanism linked to long-term-depression (LTD) (White et al. 1995b; Zhang et al. 1998; Thomas et al. 2001; Beurrier and Malenka 2002). Taken as a whole, these data are strongly consistent with the interpretation that a history of repeated cocaine is associated with decreases in excitatory synaptic and neuronal activity in the accumbens. If this were so, one would expect to observe comparable results in behaving animals with a history of cocaine self-administration. One would also expect that hypoactivity would correlate with increased propensity to seek and take drug.

The number of behaving animal recording studies that have investigated cocaine-induced plasticity is small. Nevertheless, the research that has been completed and that we will review exemplifies two approaches to testing for chronic effects of drug on neural activity in behaving animals including the following: (1) tracking the activity of individual neurons across days and (2) comparing average firing of groups of neurons recorded in animals exposed to different histories of drug self-administration. The research also exemplifies efforts to identify chronic drug effects that are related to the emergence of addiction-like patterns of behavior. Lastly, the work provides demonstrations of the methods used to differentiate chronic drug effects from normal plasticity associated with the nonpharmacological aspects of the experimental treatment.

2.8.2. Recordings in Behaving Animals

In an early behaving-animal recording study of cocaine self-administration (Peoples et al. 1999a) we tested for evidence of hypoactivity in animals with a two-week history of cocaine self-administration. In that experiment, animals self-administered an intermediate (0.75 mg/kg/inf) dose of cocaine 6 hrs per day for 14–17 days. A recording session was conducted on days 2–3 and 14–17 of training. In this experiment we used a “within-neuron” approach to testing for changes in average firing of individual neurons. We identified neurons that were recorded on both recording days. These neurons were defined as those that met the following criteria. First, the population of waveforms recorded by a single microwire had to show evidence of a minimum inter-spike interval (ISI) consistent with the refractory period of a single neuron during both sessions. Second, the waveforms recorded by the same microwire had to be comparable between sessions (stability criterion). Specifically, for both days it was necessary for neural waveforms to have been defined by the same eight discrimination parameters and by a comparable range of variation in each of those eight parameters. To be included in the between-session comparisons, neural recordings had to be consistent with an additional criterion. Specifically, the neural waveforms had to be sufficient in amplitude on each day for the smallest discriminated waveforms to exceed the amplitude of the typical maximal fluctuations in the noiseband (completeness criterion). This requirement for a minimum waveform amplitude allowed us to verify that our ability to detect discharges did not change between days 2 and 15.

On day 2 of cocaine self-administration, 83 microwires recorded activity of single neurons. Of the 83 microwires, 53 recorded neural activity on both day 2 and day 15. Of those 53 wires, only 13 (24%) yielded neural records that met the criteria of stability and completeness. The majority of the 13 neurons showed a significant decrease in basal firing between the first and the second recording session. This finding is consistent with the interpretation that self-administered cocaine induces hypoactivity, as suggested by slice and anesthetized animal studies.

We characterized the hypoactivity further in another study, which employed a between-group approach. Rats self-administered cocaine in 30, daily, long-access (6-hr) sessions. Chronic extracellular recordings of the activity of individual NAc neurons were made during sessions 2–3 and 30 of the regimen (i.e., referred to as early versus late sessions). Each recording session included a presession baseline period, a cocaine self-administration session, and a postsession recovery period. Neurons recorded on each day were categorized into two groups, task-activated and task-nonactivated. Task-activated neurons were defined as all neurons that showed either of the following firing patterns: (1) a session-long increase in firing or (2) an excitatory change in firing time-locked to a discrete task-related event, including the cues that signaled the onset of the self-administration session and the operant behavior. All other neurons were defined as task-nonactivated neurons (Peoples et al. 2007b). About half of all recorded neurons in both the early and the late sessions were task-activated, while the other half were task-nonactivated. The average firing of the two groups of neurons was compared both within each of the self-administration sessions and across the two recording sessions.

The between-session comparisons showed that the presession average firing rate of the task-nonactivated group was significantly depressed on day 30 of self-administration relative to day 2-3. In contrast, the average presession firing rate of the task-activated group did not change between the early and the late self-administration sessions (Figure 2.26). The differential between-session change in firing was associated with the emergence of a significant difference between the baseline firing rates of the task-activated and task-nonactivated neurons and an increase in the ratio of firing rates of the activated neurons relative to the nonactivated neurons. Comparable findings were observed in similar between-session comparisons of the self-administration and postsession baseline recovery phases of the early and late recording sessions. Behavior and reward-related expectations differ among the presession baseline period, the self-administration session, and the postsession baseline period of the recording sessions. The similar differential between-session changes in firing among the different phases of the recording session showed that the differential changes in firing were not attributable to between-session changes in either behavior or specific expectations about the operant session (Figure 2.27). The differential changes in firing thus potentially reflect differential drug-induced neuroadaptations.

FIGURE 2.26. Population histograms showing the average baseline and self-administration firing rates exhibited by task-activated and task-nonactivated neurons during the early and the late sessions.

FIGURE 2.26

Population histograms showing the average baseline and self-administration firing rates exhibited by task-activated and task-nonactivated neurons during the early and the late sessions. Average firing of the task-activated neurons (top) and task-nonactivated (more...)

FIGURE 2.27. Average firing of task-activated and task-nonactivated neurons: no effect of the extinction and cue reinstatement procedure.

FIGURE 2.27

Average firing of task-activated and task-nonactivated neurons: no effect of the extinction and cue reinstatement procedure. Average firing of task-activated and task-nonactivated neurons is plotted during the following periods: (1) 30 sec prior to presentation (more...)

In an additional analysis of the same data, we more finely sorted neurons based on their response characteristics. Specifically, neurons in the activated group were subdivided into two groups: (1) neurons that showed a session-long increase (referred to as session-increase neurons), and (2) neurons that showed a phasic increase in relation to a task event but showed no session-long increase (referred to as task-but-not-session-activated). Replication of the original comparisons showed that three groups of neurons exhibited different between-session changes in firing. Moreover, the between-session changes in firing exhibited by the three subtypes of neurons were positively related to the change in firing exhibited by the neurons during the self-administration session relative to the drug-free baseline period (Figure 2.25). Session-increase neurons maintained elevated rates of firing throughout individual self-administration sessions relative to presession baseline and showed a trend to increase baseline firing rates between the early and the late sessions. Task-but-not-session-activated neurons maintained average firing rates during individual self-administration sessions that were similar to baseline firing rates and correspondingly showed no between-session change in baseline. Finally, task-nonactivated neurons were inhibited during self-administration sessions and showed a between-session decrease in basal activity. The findings are consistent with the interpretation that between-session changes in firing exhibited by NAc neurons are directionally consistent with the change in firing exhibited by the neurons during individual cocaine self-administration sessions, and thus with the hypothesis that neuroadaptations induced by self-administered cocaine develop in an activity-dependent manner.

In this study, two lines of evidence supported the interpretation that the differential changes in firing of the task-activated and task-nonactivated neurons were correlated with increases in the propensity of animals to seek and take drug. First, between the early and late self-administration sessions, there was a significant increase in the average rate of cocaine self-administration exhibited by all subjects (Figure 2.28). Second, during the late recording session, the average difference in basal firing rates between task-activated and task-nonactivated neurons was predictive of animals' drug-directed behavior during the later phases of the recording session. Animals with the greatest difference in basal firing between activated and nonactivated neurons consumed the greatest amount of cocaine during the self-administration session. The same animals also completed the highest number of lever-press responses during extinction and cue-reinstatement probes that were conducted later during the recording session (Figure 2.29). The finding is suggestive of a relationship between the difference in basal firing between task-activated and task-nonactivated neurons and the propensity of animals to engage in drug-directed behavior.

FIGURE 2.28. Average rate of drug intake for all the subjects during the early and the late sessions.

FIGURE 2.28

Average rate of drug intake for all the subjects during the early and the late sessions. At the top of the figure, average interinfusion interval (i.e., min between successive self-infusions) is plotted as a function of press number. At the bottom of (more...)

FIGURE 2.29. Various indices of drug seeking and taking during the late session are compared between the high and the low-moderate drug-seekers.

FIGURE 2.29

Various indices of drug seeking and taking during the late session are compared between the high and the low-moderate drug-seekers. The bar graphs show average measures (± standard error) of the drug seeking and taking behavior exhibited by animals (more...)

2.8.3. Effects of Extended Abstinence

In our previous study (Peoples et al. 2007b) we compared phasic firing time-locked to a number of cocaine task-related events between days 2–3 and 30 of LgA (long-access, 6-hr) cocaine self-administration. At the time of each recording, animals had been exposed to a short period of drug abstinence (≈18 hrs). The task-related firing patterns that were compared included firing patterns time-locked to a cue that signaled the onset of the session and phasic firing time-locked to the cocaine-reinforced lever press. We found no significant changes in either the prevalence or the magnitude of either type of event-related phasic firing. However, in two studies, Hollander and Carelli (2005,2007) trained animals to self-administer cocaine in ShA (short-access, 2-hr) sessions and then compared task-related phasic firing patterns between animals exposed to either 1 or 30 days of abstinence. The compared firing patterns were similar to those evaluated in the Peoples et al. (2007b) study and included phasic firing time-locked to cocaine-associated cues presented to drug-free animals and operant responding during the cocaine self-administration session. Both measures of phasic activity were increased after 30 days of abstinence relative to 1 day of abstinence. Jones et al. (2008) showed that similar changes in phasic activity do not occur in animals trained to self-administer sucrose. The changes in firing observed in the Hollander and Carelli studies thus reflect cocaine-induced adaptations. The different findings in the Peoples et al. (2007b) experiment and the two Hollander and Carelli (2005,2007) studies could reflect an effect of extended abstinence, which was present in the latter studies but not in the former study, though there were additional between-experiment differences (e.g., duration of the daily self-administration session).

2.8.4. Chronic Drug-Induced Neuroadaptations: Summary, Novel Findings, and Research Opportunities

Electrophysiological recordings in slice and anesthetized animal preparations show that repeated exposure to cocaine leads to hypoactivity of NAc neurons (White 1992; White et al. 1993, 1995a; Zhang et al. 1998; Thomas et al. 2001; Beurrier and Malenka 2002). Consistent with this observation, behaving animal recordings show that repeated exposure to cocaine self-administration can lead to changes in basal firing that are consistent with the development of hypoactivity. However, those studies also show that the hypoactivity occurs only in neurons that do not exhibit excitatory changes in firing during cocaine self-administration sessions; moreover, excitatory changes in firing can be observed after extended periods of abstinence. The research also showed that the adaptations develop differentially among functionally distinct subtypes of neurons, perhaps because of the different activation patterns exhibited by those neurons during cocaine self-administration. Thus, recordings in behaving animals show that NAc firing patterns of behaving animals can undergo more complicated changes than was apparent based on the slice and anesthetized animal studies (though see Kourrich et al. 2008). The reviewed research emphasizes the additional leverage that the behaving-animal recording studies can add to electrophysiological investigations of addiction-relevant drug-induced adaptations. Indeed, characteristics of those adaptations may be subject to study only in behaving animals.

FIGURE 2.30. Average firing rates of task-activated and task-nonactivated neurons are compared between the high drug-seekers and the low-moderate drug-seekers.

FIGURE 2.30

Average firing rates of task-activated and task-nonactivated neurons are compared between the high drug-seekers and the low-moderate drug-seekers. Average firing of task-activated and task-nonactivated neurons is plotted separately for animals that were (more...)

To date there are only four studies that have characterized the effects of repeated cocaine self-administration on single neuron activity in behaving animals. The behavioral procedures used in those studies were relatively simple. Nevertheless, the studies produced some findings that are relevant to current drug addiction hypotheses. In particular, the relative and absolute increments in NAc, phasic responses to drug-associated cues (Hollander and Carelli 2005, 2007; Peoples et al. 2007b) are in line with predictions of the incentive sensitization hypothesis and the differential neuroplasticity hypothesis. The research supports the idea that behaving animal recordings can contribute to testing drug addiction hypotheses. This is likely to be especially so, as the behavioral paradigms employed in the behaving animal recording studies become more sophisticated.

2.9. FUTURE DIRECTIONS

As in any discipline, the future of behaving animal recordings for investigating drug addiction is expected to hold new technical and procedural advances. Three broad categories of advance will be discussed in this section: advances in recording technology and data analysis, advances in behavioral assays, and the combination of chronic recording with other molecular techniques. This section overviews some but not all areas of ongoing and potential advances.

2.9.1. Advances in Recording Techniques and Analysis

Chronic implantation of arrays of electrodes was not widely used before the 1980s. Advances in computing technology enabled the use of multiple-electrode arrays and have governed many other advances in the technology since then. Multiple-electrode arrays have changed greatly since their first development. Most notably, the number of electrodes in these arrays has vastly increased and will continue to increase with advances in computing technology. In addition to the increasing number of electrodes, new configurations of electrodes allow for additional information to be obtained. Electrodes with multiple recording sites, such as stereotrodes and tetrodes, allow for the identification of single neurons on multiple channels, which can increase the confidence that neural spikes originate from a single neuron (Harris et al. 2000; Buzsaki 2004).

As technology has increased the number of electrodes that can be used concurrently to record neural activity, researchers are faced with the task of discriminating individual neuron signals from those channels. The process of discriminating recorded waveforms that belong to individual neurons is a time-consuming and somewhat subjective process (see Appendix I). Better spike-sorting algorithms are expected to speed the process up, make it more objective, and increase the number of successfully identified recorded neurons (Buzsaki 2004; Schmitzer-Torbert et al. 2005; Adamos et al. 2008; Chan et al. 2008).

Large numbers of simultaneously recorded neurons can reveal information that recordings of small numbers of neurons cannot. For example, neural activity that underlies a complex behavioral sequence may be observed as a population code across multiple neurons that respond at appropriate times in the behavioral sequence. Such a population code may only become apparent when looking at recordings of large numbers of neurons. Investigators are increasingly considering neurons as participants in networks of activity, and using analyses that observe the dynamics of large numbers of neurons simultaneously, rather than analyzing individual neurons independently (Petersen et al. 2002; Panzeri et al. 2003; Sanger 2003; Mazor and Laurent 2005; Averbeck et al. 2006; Fontanini and Katz 2006; Reddy and Kanwisher 2006; Lemon and Katz 2007).

2.9.2. Advances in Behavioral Assays

Addiction is thought to include multiple stages, including controlled drug use, drug abuse, compulsive and uncontrolled drug seeking, abstinence, and relapse. An effort to extend the recording technique to self-administration paradigms that model these different phases will help to identify chronic drug-induced changes in neuronal activity that contribute to the progression of the disorder. The different findings of the Peoples et al. (2007) and Hollander and Carelli (2005, 2007) studies suggest that neurons in regions implicated in addiction, such as the NAc, may undergo a series of neuroadaptive changes in activity across phases of active drug use and extended abstinence. A recent nicotine self-administration study in our laboratory (K Guillem and LL Peoples, in press) supports this idea. Characterizing neural activity across phases of initial drug use, abstinence, and re-exposure could be very important to a full understanding of changes in neuronal activity that contribute to drug addiction (Guillem and Peoples, in press).

Greater application of the chronic recording procedure to studies of drug seeking would extend the utility of the method. Relapse prevention is a primary therapeutic endpoint in addiction treatment. As we have already described, there is evidence that drug-directed behavior under drug-free and drug-exposed conditions is mediated by somewhat different mechanisms. The multiphase seeking-taking procedure that we are developing and that was described in this chapter may be a useful paradigm for characterizing neural mechanisms associated with drug seeking under drug-free conditions. The method may be preferable relative to the common employed extinction-reinstatement procedures, which do not mimic conditions of abstinence and relapse in humans (i.e., humans do not undergo instrumental extinction). Application of the behaving-animal recording method to studies of chronic drug effects could be greatly advanced by these experimental design and procedural developments.

Progress in applying the chronic-behaving-animal recording procedure to investigating drug addiction hypotheses will depend to some degree on inclusion of more sophisticated behavioral approaches to the recording experiments. For example, procedures that temporally isolate events or motivational processes implicated by addiction hypotheses will be necessary if we are to identify the correlated neural responses and to test the effects of repeated drug exposure on those firing patterns. Studies of natural reward-directed behavior in animals with various histories of drug self-administration and abstinence will also be necessary additions to behaving animal studies of chronic drug effects that would be helpful in testing certain hypotheses. Finally, a variety of behavioral paradigms have been developed to test certain of the cognitive and motivational processes hypothesized to be affected by drug-induced adaptations and to be involved in addiction. Incorporating these paradigms in chronic animal recording studies would increase the power of those investigations to test addiction hypotheses.

2.9.3. Getting at Mechanism: Combining Chronic Recording with Other Techniques

The identification of activity patterns of single neurons and groups of neurons that are related to drug seeking and drug reward is an important contribution of the chronic recording studies. However, the power of the research will be greatly enhanced if we can develop the methodology to more readily investigate the mechanisms that mediate those neural firing patterns. To do so it will be necessary to integrate the recording technique with additional neurochemical, neuropharmacological, biochemical, and molecular techniques.

Combining chronic recordings with molecular and cellular techniques, such as measurement of change in protein levels and receptor expression, will be useful for investigating the mechanisms of drug action. These studies most likely could best be conducted in separate sets of animals matched to the recording animals for drug exposure and so forth. The parallel studies could be useful in testing whether treatment regimens that engender certain changes in neural activity also engender particular biochemical and molecular adaptations that have been proposed to contribute to addiction.

Recent advances in molecular intervention methods using antisense and viruses now provide the unique opportunity to alter receptor functionality in a brain, region-specific manner. Application of these treatments at the time of surgery, could allow researchers to test specific hypotheses about receptor mechanisms and signal transduction pathways that might mediate acute and chronic effects of drug on neural activity.

Some of these suggested future directions are easier to implement than others. However, integration of the chronic recordings with the additional techniques will provide the opportunity to bridge the functional behaving-animal recording studies with other neuroscience fields of drug addiction. This could contribute significantly to developing a more complete picture of the mechanisms that mediate reward and addiction.

2.10. CONCLUSIONS

In the last 20–30 years, researchers have investigated the neurophysiological effects of acute and repeated exposure to self-administered drug. Much of the completed work has been conducted in slice and anesthetized animal preparations. Although these procedures can be used to conduct elegant mechanistic studies of acute and chronic drug effects on single neuron activity, drug actions and effects in these preparations do not always correspond with those that occur in behaving animals. This is because anesthesia and the absence of normal afferent input to neurons during the recording procedures can alter drug actions and effects. To address these issues it is necessary to characterize drug effects in behaving animals with a relevant history of drug self-administration.

The chronic-behaving-animal recording procedure can be used to characterize both acute and chronic effects of self-administered drug on individual neuron firing and to identify those particular drug-induced changes in neural activity that are associated with the emergence of addiction-like patterns of behavior. The method can also be used to test drug addiction hypotheses, which generally propose that drug-induced alterations in neural responses associated with specific motivational processes underlie the emergence of addiction-like behaviors. Though the chronic-behaving-animal recording method affords a number of novel research opportunities, the utility of the procedure in identifying pharmacological effects of addictive drugs depends on certain behavioral controls. Specifically, use of the chronic recording method to investigate acute pharmacological effects requires controls for afferent feedback associated with drug-induced changes in behavior and motivational processes. Similarly, use of the method in studies of chronic drug-induced adaptations requires controls for normal experiential plasticity that might occur with repeated exposure to the experimental procedures. Application of the technique to test addiction hypotheses will also depend on attention to various behavioral issues.

Thus far, the chronic-behaving-animal recording procedure has been applied primarily to studies of accumbal neural activity in animals trained to self-administer cocaine. A portion of those studies were reviewed in the present chapter. The research that was described exemplifies some of the methods that can be used to identify functional correlates of single neuron activity during drug seeking and taking, and to investigate acute and chronic effects of self-administered drug. The research that was described also highlighted a number of novel findings that have substantive implications for understanding mechanisms that mediate cocaine-directed behavior and cocaine addiction. Some of these include the following. First, task-related firing patterns of NAc neurons appear to be similar during self-administration of cocaine and natural, food, or fluid rewards. However, the neurons that exhibit the task-related firing patterns differ depending on the reward. Second, drug-associated cues and cocaine-directed behavior are associated with predominantly excitatory phasic (event-related) NAc activity under both drug-free and drug-exposed conditions. Both drug seeking and taking are also associated with sustained changes in average firing. These changes are almost exclusively excitatory in the absence of reinforcement. However, increases and decreases in firing occur when animals are exposed to conditioned and primary cocaine reinforcement, with decreases actually predominating during cocaine self-administration. Third, self-administered cocaine has primarily inhibitory acute effects on NAc activity, and repeated exposure to cocaine self-administration can decrease NAc basal activity. These findings are consistent with those of slice and anesthetized animal recording studies. However, the chronic recording studies showed that acute and chronic cocaine effects are more complex in behaving animals as compared with slice and anesthetized animal preparations. Specifically, in behaving animals, neurons exhibit heterogeneous and activity-dependent responses to acute and chronic drug exposure. In fact, some neurons exhibit excitatory rather than inhibitory changes in firing. Finally, chronic recording studies have shown that NAc neurons exhibit changes in firing in response to repeated cocaine self-administration that are consistent with predictions of the incentive sensitization hypothesis and the differential neuroplasticity hypothesis. These findings demonstrate the utility of the chronic behaving animal recording method. Indeed there may be mechanisms that are critical to addiction which are only observable in the behaving animal studies.

Despite the advantages of the chronic-behaving-animal recording method, the procedure is currently not conducive to the same types of mechanistic studies that can be employed in acute and anesthetized electrophysiological studies of drug actions. Given the relative strengths and weaknesses of the different electrophysiological methods, complementary application of the techniques and consideration of findings from those studies as a whole will be important to understanding the drug effects that contribute to drug addiction. Slice and anesthetized animal methods can be used to identify drug effects and to characterize mechanisms of drug action. Recordings in behaving animals can be used to test for changes in neuronal activity that are predicted on the basis of the slice and anesthetized recordings, to identify additional drug effects that may be critical to addiction but absent in the typical slice and anesthetized animal studies, to test for changes in neural activity that are linked to particular addiction-like patterns of behavior, and to test certain predictions of drug addiction hypotheses. The advantages that can be gained by an integrative approach are demonstrated by the research described in the present chapter.

ACKNOWLEDGMENTS

LLP supported by NIH/NIDA P60 DA 005186 (CP O'Brien) and NIH/NIDA P50 DA 012756 (H Pettinati). KG supported by NIH Director's Bench-to-Bedside (LLP, Elliot Stein). AVK supported by DA-07241(CP O'Brien) and DA-021449 (AVK).

REFERENCES

  • Adamos DA, Kosmidis EK, Theophilidis G. Performance evaluation of PCA-based spike sorting algorithms. Comput Methods Programs Biomed. 2008;91:232–244. [PubMed: 18565614]
  • Ahmed SH, Koob GF. Transition from moderate to excessive drug intake: change in hedonic set point. Science. 1998;282:298–300. [PubMed: 9765157]
  • American Psychiatric Association. Diagnostic and statistical manual of mental disorders. (4th ed) Washington, D.C.: American Psychiatric Association; 1994.
  • Antkowiak B. Different actions of general anesthetics on the firing patterns of neocortical neurons mediated by the GABA(A) receptor. Anesthesiology. 1999;91:500–511. [PubMed: 10443614]
  • Antkowiak B, Helfrich-Forster C. Effects of small concentrations of volatile anesthetics on action potential firing of neocortical neurons in vitro. Anesthesiology. 1998;88:1592–1605. [PubMed: 9637654]
  • Apicella P, Ljungberg T, Scarnati E, Schultz W. Responses to reward in monkey dorsal and ventral striatum. Exp Brain Res. 1991;85:491–500. [PubMed: 1915708]
  • Arroyo M, Markou A, Robbins TW, Everitt BJ. Acquisition, maintenance and reinstatement of intravenous cocaine self-administration under a second-order schedule of reinforcement in rats: effects of conditioned cues and continuous access to cocaine. Psychopharmacology (Berl) 1998;140:331–344. [PubMed: 9877013]
  • Averbeck BB, Latham PE, Pouget A. Neural correlations, population coding and computation. Nat Rev Neurosci. 2006;7:358–366. [PubMed: 16760916]
  • Ben-Shahar O, Keeley P, Cook M, Brake W, Joyce M, Nyffeler M, Heston R, Ettenberg A. Changes in levels of Dl, D2, or NMDA receptors during withdrawal from brief or extended daily access to IV cocaine. Brain Res. 2007;1131:220–228. [PMC free article: PMC1800943] [PubMed: 17161392]
  • Beninger RJ. The role of dopamine in locomotor activity and learning. Brain Res. 1983;287:173–196. [PubMed: 6357357]
  • Berke JD, Hyman SE. Addiction, dopamine, and the molecular mechanisms of memory. Neuron. 2000;25:515–532. [PubMed: 10774721]
  • Beurrier C, Malenka RC. Enhanced inhibition of synaptic transmission by dopamine in the nucleus accumbens during behavioral sensitization to cocaine. J Neurosci. 2002;22:5817–5822. [PubMed: 12122043]
  • Bowman EM, Aigner TG, Richmond BJ. Neural signals in the monkey ventral striatum related to motivation for juice and cocaine rewards. J Neurophysiol. 1996;75:1061–1073. [PubMed: 8867118]
  • Breiter HC, Gollub RL, Weisskoff RM, Kennedy DN, Makris N, Berke JD, Goodman JM. Acute effects of cocaine on human brain activity and emotion. Neuron. 1997;19:591–611. [PubMed: 9331351]
  • Buzsaki G. Large-scale recording of neuronal ensembles. Nat Neurosci. 2004;7:446–451. [PubMed: 15114356]
  • Cador M, Robbins TW, Everitt BJ. Involvement of the amygdala in stimulus-reward associations: interaction with the ventral striatum. Neuroscience. 1989;30:77–86. [PubMed: 2664556]
  • Cador M, Taylor JR, Robbins TW. Potentiation of the effects of reward-related stimuli by dopaminergic-dependent mechanisms in the nucleus accumbens. Psychopharmacology (Berl) 1991;104:377–385. [PubMed: 1924645]
  • Calu DJ, Roesch MR, Stalnaker TA, Schoenbaum G. Associative encoding in posterior piriform cortex during odor discrimination and reversal learning. Cereb Cortex. 2007;17:1342–1349. [PMC free article: PMC2473864] [PubMed: 16882682]
  • Cardinal RN, Parkinson JA, Lachenal G, Halkerston KM, Rudarakanchana N, Hall J, Morrison CH, Howes SR, Robbins TW, Everitt BJ. Effects of selective excitotoxic lesions of the nucleus accumbens core, anterior cingulate cortex, and central nucleus of the amygdala on autoshaping performance in rats. Behav Neurosci. 2002;116:553–567. [PubMed: 12148923]
  • Carelli RM. Activation of accumbens cell firing by stimuli associated with cocaine delivery during self-administration. Synapse. 2000;35:238–242. [PubMed: 10657032]
  • Carelli RM, Ijames SG. Nucleus accumbens cell firing during maintenance, extinction, and reinstatement of cocaine self-administration behavior in rats. Brain Res. 2000;866:44–54. [PubMed: 10825479]
  • Carelli RM, Ijames SG, Crumling AJ. Evidence that separate neural circuits in the nucleus accumbens encode cocaine versus “natural” (water and food) reward. J Neurosci. 2000;20:4255–4266. [PubMed: 10818162]
  • Carelli RM, King VC, Hampson RE, Deadwyler SA. Firing patterns of nucleus accumbens neurons during cocaine self-administration in rats. Brain Res. 1993;626:14–22. [PubMed: 8281424]
  • Carelli RM, Wightman RM. Functional microcircuitry in the accumbens underlying drug addiction: insights from real-time signaling during behavior. Curr Opin Neurobiol. 2004;14:763–768. [PubMed: 15582381]
  • Chan HL, Wu T, Lee ST, Fang SC, Chao PK, Lin MA. Classification of neuronal spikes over the reconstructed phase space. J Neurosci Methods. 2008;168:203–211. [PubMed: 17976735]
  • Chang JY, Janak PH, Woodward DJ. Comparison of mesocorticolimbic neuronal responses during cocaine and heroin self-administration in freely moving rats. J Neurosci. 1998;18:3098–3115. [PubMed: 9526026]
  • Chang JY, Sawyer SF, Lee RS, Woodward DJ. Electrophysiological and pharmacological evidence for the role of the nucleus accumbens in cocaine self-administration in freely moving rats. J Neurosci. 1994;14:1224–1244. [PubMed: 8120621]
  • Childress AR, Hole AV, Ehrman RN, Robbins SJ, Mc Lellan AT, O’Brien CP. Cue reactivity and cue reactivity interventions in drug dependence. NIDA Res Monogr. 1993;137:73–95. [PubMed: 8289929]
  • Childress AR, Mozley PD, Mc Elgin W, Fitzgerald J, Reivich M, O’Brien CP. Limbic activation during cue-induced cocaine craving. Am J Psychiatry. 1999;156:11–18. [PMC free article: PMC2820826] [PubMed: 9892292]
  • Colpaert FC. Drug discrimination: methods of manipulation, measurement, and analysis. In: Bozarth M.A., editor. Methods of assessing the reinforcing properties of abused drugs. New York: Springer; 1987. pp. 341–372.
  • Cromwell HC, Schultz W. Effects of expectations for different reward magnitudes on neuronal activity in primate striatum. J Neurophysiol. 2003;89:2823–2838. [PubMed: 12611937]
  • Deadwyler SA. Electrophysiological investigations of drug influences in the behaving animal. In: Geller H.M., editor. Modern methods in pharmacology Vol. 3 Electrophysiological techniques in pharmacology: 1986. Chap 1.
  • Deroche-Gamonet V, Belin D, Piazza PV. Evidence for addiction-like behavior in the rat. Science. 2004;305:1014–1017. [PubMed: 15310906]
  • Di Chiara G, Acquas E, Carboni E. Drug motivation and abuse: a neurobiological perspective. Ann N Y Acad Sci. 1992;654:207–219. [PubMed: 1632584]
  • Dickinson A, Wood N, Smith JW. Alcohol seeking by rats: action or habit? Q J Exp Psychol B. 2002;55:331–348. [PubMed: 12350285]
  • Everitt BJ, Cador M, Robbins TW. Interactions between the amygdala and ventral striatum in stimulus-reward associations: studies using a second-order schedule of sexual reinforcement. Neuroscience. 1989;30:63–75. [PubMed: 2664555]
  • Everitt BJ, Dickinson A, Robbins TW. The neuropsychological basis of addictive behaviour. Brain Res Brain Res Rev. 2001;36:129–138. [PubMed: 11690609]
  • Everitt BJ, Wolf ME. Psychomotor stimulant addiction: a neural systems perspective. J Neurosci. 2002;22:3312–3320. [PubMed: 11978805]
  • Fink-Jensen A, Ingwersen SH, Nielsen PG, Hansen L, Nielsen EB, Hansen AJ. Halothane anesthesia enhances the effect of dopamine uptake inhibition on interstitial levels of striatal dopamine. Naunyn Schmiedebergs Arch Pharmacol. 1994;350:239–244. [PubMed: 7824039]
  • Floresco SB, West AR, Ash B, Moore H, Grace AA. Afferent modulation of dopamine neuron firing differentially regulates tonic and phasic dopamine transmission. Nat Neurosci. 2003;6:968–973. [PubMed: 12897785]
  • Fontanini A, Katz DB. State-dependent modulation of time-varying gustatory responses. J Neurophysiol. 2006;96:3183–3193. [PubMed: 16928791]
  • Fuchs RA, Branham RK, See RE. Different neural substrates mediate cocaine seeking after abstinence versus extinction training: a critical role for the dorsolateral caudate-putamen. J Neurosci. 2006;26:3584–3588. [PMC free article: PMC1643847] [PubMed: 16571766]
  • Garavan H, Pankiewicz J, Bloom A, Cho JK, Sperry L, Ross TJ, Salmeron BJ, Risinger R, Kelley D, Stein EA. Cue-induced cocaine craving: neuroanatomical specificity for drug users and drug stimuli. Am J Psychiatry. 2000;157:1789–1798. [PubMed: 11058476]
  • Gawin FH, Kleber HD. Abstinence symptomatology and psychiatric diagnosis in cocaine abusers. Clinical observations. Arch Gen Psychiatry. 1986;43:107–113. [PubMed: 3947206]
  • Ghitza UE, Fabbricatore AT, Prokopenko V, Pawlak AP, West MO. Persistent cue-evoked activity of accumbens neurons after prolonged abstinence from self-administered cocaine. J Neurosci. 2003;23:7239–7245. [PubMed: 12917356]
  • Grant S, London ED, Newlin DB, Villemagne VL, Liu X, Contoreggi C, Phillips RL, Kimes AS, Margolin A. Activation of memory circuits during cue-elicited cocaine craving. Proc Natl Acad Sci U S A. 1996;93:12040–12045. [PMC free article: PMC38179] [PubMed: 8876259]
  • Green JD. A simple microelectrode for recording from the central nervous system. Nature. 1958;182:962. [PubMed: 13590200]
  • Griffiths RR, Bigelow GE, Henningfield JE. Animal and human drug-taking behavior. In: Mello NK, editor. Advances in Substance Abuse Behavioral and Biological Research. JAI Press; Greenwich, CT: 1980. Chap. 3.
  • Grimm JW, See RE. Dissociation of primary and secondary reward-relevant limbic nuclei in an animal model of relapse. Neuropsychopharmacology. 2000;22:473–479. [PubMed: 10731622]
  • Gullienk K, Peoples LL. Progressive and lasting amplification of accumbal nicotine-seeking neural signals. J Neurosci. [PMC free article: PMC2855140] [PubMed: 20053909]
  • Harris KD, Henze DA, Csicsvari J, Hirase H, Buzsaki G. Accuracy of tetrode spike separation as determined by simultaneous intracellular and extracellular measurements. J Neurophysiol. 2000;84:401–414. [PubMed: 10899214]
  • Hassani OK, Cromwell HC, Schultz W. Influence of expectation of different rewards on behavior-related neuronal activity in the striatum. J Neurophysiol. 2001;85:2477–2489. [PubMed: 11387394]
  • Hemby SE, Co C, Koves TR, Smith JE, Dworkin SI. Differences in extracellular dopamine concentrations in the nucleus accumbens during response-dependent and response-independent cocaine administration in the rat. Psychopharmacology (Berl) 1997;133:7–16. [PubMed: 9335075]
  • Henry DJ, White FJ. The persistence of behavioral sensitization to cocaine parallels enhanced inhibition of nucleus accumbens neurons. J Neurosci. 1995;15:6287–6299. [PubMed: 7666211]
  • Hernandez-Lopez S, Bargas J, Surmeier DJ, Reyes A, Galarraga E. Dl receptor activation enhances evoked discharge in neostriatal medium spiny neurons by modulating an L-type Ca2+ conductance. J Neurosci. 1997;17:3334–3342. [PubMed: 9096166]
  • Hollander JA, Carelli RM. Abstinence from cocaine self-administration heightens neural encoding of goal-directed behaviors in the accumbens. Neuropsychopharmacology. 2005;30:1464–1474. [PubMed: 15856078]
  • Hollander JA, Carelli RM. Cocaine-associated stimuli increase cocaine seeking and activate accumbens core neurons after abstinence. J Neurosci. 2007;27:3535–3539. [PubMed: 17392469]
  • Hollerman JR, Tremblay L, Schultz W. Influence of reward expectation on behavior-related neuronal activity in primate striatum. J Neurophysiol. 1998;80:947–963. [PubMed: 9705481]
  • Hope BT, Crombag HS, Jedynak JP, Wise RA. Neuroadaptations of total levels of adenylate cyclase, protein kinase A, tyrosine hydroxylase, cdk5 and neurofilaments in the nucleus accumbens and ventral tegmental area do not correlate with expression of sensitized or tolerant locomotor responses to cocaine. J Neurochem. 2005;92:536–545. [PubMed: 15659224]
  • Hu XT, Basu S, White FJ. Repeated cocaine administration suppresses HVA-Ca2+ potentials and enhances activity of K+ channels in rat nucleus accumbens neurons. J Neurophysiol. 2004;92:1597–1607. [PubMed: 15331648]
  • Hu XT, Ford K, White FJ. Repeated cocaine administration decreases calcineurin (PP2B) but enhances DARPP-32 modulation of sodium currents in rat nucleus accumbens neurons. Neuropsychopharmacology. 2005;30:916–926. [PubMed: 15726118]
  • Hu XT, White FJ. Loss of D1/D2 dopamine receptor synergisms following repeated administration of Dl or D2 receptor selective antagonists: electrophysiological and behavioral studies. Synapse. 1994;17:43–61. [PubMed: 7913772]
  • Hurd YL, Kehr J, Ungerstedt U. In vivo microdialysis as a technique to monitor drug transport: correlation of extracellular cocaine levels and dopamine overflow in the rat brain. J Neurochem. 1988;51:1314–1316. [PubMed: 3418351]
  • Ito R, Dalley JW, Robbins TW, Everitt BJ. Dopamine release in the dorsal striatum during cocaine-seeking behavior under the control of a drug-associated cue. J Neurosci. 2002;22:6247–6253. [PubMed: 12122083]
  • Ito R, Dalley JW, Howes SR, Robbins TW, Everitt BJ. Dissociation in conditioned dopamine release in the nucleus accumbens core and shell in response to cocaine cues and during cocaine-seeking behavior in rats. J Neurosci. 2000;20:7489–7495. [PubMed: 11007908]
  • Iversen SD, Koob GF. Behavioral implications of dopaminergic neurons in the mesolim-bic system. Adv Biochem Psychopharmacol. 1977;16:209–214. [PubMed: 560790]
  • Janak PH. Multichannel neural ensemble recording during alcohol self-administration. In: Liu Y., Lovinger D.M., editors. Methods for alcohol-related neuroscience research. CRC Press; Boca Raton, FL: 2002. pp. 243–259.
  • Janak PH. Application of many-neuron microelectrode array recording and the study of reward seeking behavior. In: Waterhouse BD, editor. Methods in drug abuse research: cellular and circuit level analyses. CRC Press; Boca Raton, FL: 2003.
  • Johanson CE, Balster RL. A summary of the results of a drug self-administration study using substitution procedures in rhesus monkeys. Bull Narc. 1978;30:43–54. [PubMed: 36945]
  • Johanson CE, Schuster CR. A choice procedure for drug reinforcers: cocaine and methylphenidate in the rhesus monkey. J Pharmacol Exp Ther. 1975;193:676–688. [PubMed: 1142112]
  • Jones JL, Wheeler RA, Carelli RM. Behavioral responding and nucleus accumbens cell firing are unaltered following periods of abstinence from sucrose. Synapse. 2008;62:219–228. [PubMed: 18088061]
  • Kalivas PW. Recent understanding in the mechanisms of addiction. Curr Psychiatry Rep. 2004;6:347–351. [PubMed: 15355757]
  • Kalivas PW, O’Brien C. Drug addiction as a pathology of staged neuroplasticity. Neuropsychopharmacology. 2008;33:166–180. [PubMed: 17805308]
  • Kelley AE, Stinus L. Disappearance of hoarding behavior after 6-hydroxydopamine lesions of the mesolimbic dopamine neurons and its reinstatement with L-dopa. Behav Neurosci. 1985;99:531–545. [PubMed: 3939664]
  • Kiyatkin EA, Rebec GV. Dopaminergic modulation of glutamate-induced excitations of neurons in the neostriatum and nucleus accumbens of awake, unrestrained rats. J Neurophysiol. 1996;75:142–153. [PubMed: 8822548]
  • Koob GF, Le Moal M. Drug abuse: hedonic homeostatic dysregulation. Science. 1997;278:52–58. [PubMed: 9311926]
  • Koob GF, Le Moal M. Drug addiction, dysregulation of reward, and allostasis. Neuro-psychopharmacology. 2001;24:97–129. [PubMed: 11120394]
  • Koob GF, Le Moal M. Addiction and the brain antireward system. Annu Rev Psychol. 2008;59:29–53. [PubMed: 18154498]
  • Koob GF, Nestler EJ. The neurobiology of drug addiction. Neuropsychiatry Clin Neurosci. 1997;9:482–497. [PubMed: 9276849]
  • Koob GF, Stinus L, Le Moal M, Bloom FE. Opponent process theory of motivation: neurobiological evidence from studies of opiate dependence. Neurosci Biobehav Rev. 1989;13:135–140. [PubMed: 2682399]
  • Kourrich S, Rothwell PE, Klug JR, Thomas MJ. Cocaine experience controls bidirectional synaptic plasticity in the nucleus accumbens. J Neurosci. 2007;27:7921–7928. [PubMed: 17652583]
  • Kravitz AV, Moorman DE, Simpson A, Peoples LL. Session-long modulations of accumbal firing during sucrose-reinforced operant behavior. Synapse. 2006;60:420–428. [PubMed: 16881071]
  • Kravitz AV, Peoples LL. Background firing rates of orbitofrontal neurons reflect specific characteristics of operant sessions and modulate phasic responses to reward-associated cues and behavior. J Neurosci. 2008;28:1009–1018. [PubMed: 18216208]
  • Lavoie AM, Mizumori SJ. Spatial, movement- and reward-sensitive discharge by medial ventral striatum neurons of rats. Brain Res. 1994;638:157–168. [PubMed: 8199856]
  • Lemon CH, Katz DB. The neural processing of taste. BMC Neurosci. 2007;8 Suppl:3–S5. [PMC free article: PMC1995451] [PubMed: 17903281]
  • Levine MS, Li Z, Cepeda C, Cromwell HC, Altemus KL. Neuromodulatory actions of dopamine on synaptically-evoked neostriatal responses in slices. Synapse. 1996;24:65–78. [PubMed: 9046078]
  • Lu L, Dempsey J, Shaham Y, Hope BT. Differential long-term neuroadaptations of glutamate receptors in the basolateral and central amygdala after withdrawal from cocaine self-administration in rats. J Neurochem. 2005;94:161–168. [PubMed: 15953359]
  • Mantsch JR, Ho A, Schlussman SD, Kreek MJ. Predictable individual differences in the initiation of cocaine self-administration by rats under extended-access conditions are dose-dependent. Psychopharmacology (Berl) 2001;157:31–39. [PubMed: 11512040]
  • Markou A, Arroyo M, Everitt BJ. Effects of contingent and non-contingent cocaine on drug-seeking behavior measured using a second-order schedule of cocaine reinforcement in rats. Neuropsychopharmacology. 1999;20:542–555. [PubMed: 10327424]
  • Mazor O, Laurent G. Transient dynamics versus fixed points in odor representations by locust antennal lobe projection neurons. Neuron. 2005;48:661–673. [PubMed: 16301181]
  • Mereu G, Casu M, Gessa GL. Sulpiride activates the firing rate and tyrosine hydroxylase activity of dopaminergic neurons in unanesthetized rats. Brain Res. 1983;264:105–110. [PubMed: 6133578]
  • Misra AL, Pontani RB, Mule SJ. [3H]-Noncocaine and [3H]-pseudococaine: effect of N-demethylation and C2-epimerization of cocaine on its pharmacokinetics in the rat. Experientia. 1976;32:895–897. [PubMed: 954975]
  • Morgenson G, Yim C. Neuromodulatory functions of the mesolimbic dopamine system: electrophysiological and behavioral studies. In: Willner P, Scheel-Kruger J, editors. The mesolimbic dopamine system: from motivation to action. New York: Wiley; 1991.
  • Moxon KA. Multichannel electrode design: considerations for different application. In: Nicolelis M.A.L., editor. Methods for Neural Ensemble Recordings. Boca Raton, FL: CRC Press; 1999.
  • Nasif FJ, Hu XT, White FJ. Repeated cocaine administration increases voltage-sensitive calcium currents in response to membrane depolarization in medial prefrontal cortex pyramidal neurons. J Neurosci. 2005;25:3674–3679. [PubMed: 15814798]
  • Nestler EJ. Molecular basis of long-term plasticity underlying addiction. Nat Rev Neurosci. 2001;2:119–128. [PubMed: 11252991]
  • Nicola SM, Kombian SB, Malenka RC. Psychostimulants depress excitatory synaptic transmission in the nucleus accumbens via presynaptic Dl-like dopamine receptors. J Neurosci. 1996;16:1591–1604. [PubMed: 8774428]
  • Nicola SM, Surmeier J, Malenka RC. Dopaminergic modulation of neuronal excitability in the striatum and nucleus accumbens. Annu Rev Neurosci. 2000;23:185–215. [PubMed: 10845063]
  • Nicola SM, Yun IA, Wakabayashi KT, Fields HL. Firing of nucleus accumbens neurons during the consummatory phase of a discriminative stimulus task depends on previous reward predictive cues. J Neurophysiol. 2004;91:1866–1882. [PubMed: 14645378]
  • O’Brien CP. Anticraving medications for relapse prevention: a possible new class of psychoactive medications. Am J Psychiatry. 2005;162:1423–1431. [PubMed: 16055763]
  • O’Brien CP, Childress AR, Mc Lellan T, Ehrman R. Integrating systemic cue exposure with standard treatment in recovering drug dependent patients. Addict Behav. 1990;15:355–365. [PubMed: 2248109]
  • O’Brien CP, Childress AR, Ehrman R, Robbins SJ. Conditioning factors in drug abuse: can they explain compulsion? J Psychopharmacol. 1998;12:15–22. [PubMed: 9584964]
  • O’Donnell P. Dopamine gating of forebrain neural ensembles. Eur J Neurosci. 2003;17:429–135. [PubMed: 12581161]
  • O’Donnell P, Grace AA. Dopaminergic reduction of excitability in nucleus accumbens neurons recorded in vitro. Neuropsychopharmacology. 1996;15:87–97. [PubMed: 8797195]
  • O’Donnell P, Greene J, Pabello N, Lewis BL, Grace AA. Modulation of cell firing in the nucleus accumbens. Ann NY Acad Sci. 1999;877:157–175. [PubMed: 10415649]
  • Olmstead MC, Lafond MV, Everitt BJ, Dickinson A. Cocaine seeking by rats is a goal-directed action. Behav Neurosci. 2001;115:394–402. [PubMed: 11345964]
  • Olmstead MC, Parkinson JA, Miles FJ, Everitt BJ, Dickinson A. Cocaine-seeking by rats: regulation, reinforcement and activation. Psychopharmacology (Berl) 2000;152:123–131. [PubMed: 11057515]
  • Overton DA. Applications and limitations of the drug discrimination method for the study of drug abuse. In: Bozarth M.A., editor. Methods of assession the reinforcing properties of abused drugs. New York: Springer; 1987. pp. pp 291–340.
  • Panlilio LV, Goldberg SR. Self-administration of drugs in animals and humans as a model and an investigative tool. Addiction. 2007;102:1863–1870. [PMC free article: PMC2695138] [PubMed: 18031422]
  • Panzeri S, Pola G, Petersen RS. Coding of sensory signals by neuronal populations: the role of correlated activity. Neuroscientist. 2003;9:175–180. [PubMed: 15065813]
  • Paxinos G, Watson C. The rat brain in stereotaxic coordinates. 4th Ed. New York: Elsevier; 2004.
  • Peltier RL, Guerin GF, Dorairaj N, Goeders NE. Effects of saline substitution on responding and plasma corticosterone in rats trained to self-administer different doses of cocaine. J Pharmacol Exp Ther. 2001;299:114–120. [PubMed: 11561070]
  • Pennartz CM, Groenewegen HJ, Lopes da, Silva FH. The nucleus accumbens as a complex of functionally distinct neuronal ensembles: an integration of behavioural, electrophysiological and anatomical data. Prog Neurobiol. 1994;42:719–761. [PubMed: 7938546]
  • Peoples LL. Application of chronic extracellular recording to studies of drug self-administration. In: Waterhouse B.D., editor. Methods in drug abuse research: cellular and circuits level analyses. Boca Raton, FL: CRC Press; 2003. pp. 161–211.
  • Peoples LL, Bibi R, West MO. Effects of intravenous selfadministered cocaine on single cell activity in the nucleus accumbens of the rat. In: Harris L, editor. Problems of drug dependence, 1993: proceedings of the 55th annual scientific meeting. The College on Problems of Drug Dependence, Inc. II. Washington, D.C: Superintendent of Documents, United States Government Printing Office; 1994. p. 326. National Institute on Drug Abuse Research Monograph 141.
  • Peoples LL, Cavanaugh D. Differential changes in signal and background firing of accumbal neurons during cocaine self-administration. J Neurophysiol. 2003;90:993–1010. [PubMed: 12904500]
  • Peoples LL, Gee F, Bibi R, West MO. Phasic firing time locked to cocaine self-infusion and locomotion: dissociable firing patterns of single nucleus accumbens neurons in the rat. J Neurosci. (1998b);18:7588–7598. [PubMed: 9736676]
  • Peoples LL, Kravitz AV, Guillem K. The role of accumbal hypoactivity in cocaine addiction. ScientificWorldJournal. (2007a);7:22–45. [PubMed: 17982574]
  • Peoples LL, Kravitz AV, Lynch KG, Cavanaugh DJ. Accumbal neurons that are activated during cocaine self-administration are spared from inhibitory effects of repeated cocaine self-administration. Neuropsychopharmacology. (2007b);32:1141–1158. [PubMed: 17019407]
  • Peoples LL, Lynch KG, Lesnock J, Gangadhar N. Accumbal neural responses during the initiation and maintenance of intravenous cocaine self-administration. Neurophysiol. 2004;91:314–323. [PubMed: 14523071]
  • Peoples LL, Uzwiak AJ, Gee F, Fabbricatore AT, Muccino KJ, Mohta BD, West MO. Phasic accumbal firing may contribute to the regulation of drug taking during intravenous cocaine self-administration sessions. Ann NY Acad Sci. (1999b);877:781–787. [PubMed: 10415704]
  • Peoples LL, Uzwiak AJ, Gee F, West MO. Operant behavior during sessions of intravenous cocaine infusion is necessary and sufficient for phasic firing of single nucleus accumbens neurons. Brain Res. 1997;757:280–284. [PubMed: 9200758]
  • Peoples LL, Uzwiak AJ, Gee F, West MO. Tonic firing of rat nucleus accumbens neurons: changes during the first 2 weeks of daily cocaine self-administration sessions. Brain Res. (1999a);822:231–236. [PubMed: 10082901]
  • Peoples LL, Uzwiak AJ, Guyette FX, West MO. Tonic inhibition of single nucleus accumbens neurons in the rat: a predominant but not exclusive firing pattern induced by cocaine self-administration sessions. Neuroscience. (1998a);86:13–22. [PubMed: 9692739]
  • Peoples LL, West MO. Phasic firing of single neurons in the rat nucleus accumbens correlated with the timing of intravenous cocaine self-administration. Neurosci. 1996;16:3459–3473. [PubMed: 8627379]
  • Petersen RS, Panzeri S, Diamond ME. Population coding in somatosensory cortex. Curr Opin Neurobiol. 2002;12:441–447. [PubMed: 12139993]
  • Pettit HO, Justice JB Jr. Dopamine in the nucleus accumbens during cocaine self-administration as studied by in vivo microdialysis. Pharmacol Biochem Behav. 1989;34:899–904. [PubMed: 2623043]
  • Pickens R, Thompson T. Cocaine-reinforced behavior in rats: effects of reinforcement magnitude and fixed-ratio size. J Pharmacol Exp Ther. 1968;161:122–129. [PubMed: 5648489]
  • Pierce RC, Rebec GV. Iontophoresis in the neostriatum of awake, unrestrained rats: differential effects of dopamine, glutamate and ascorbate on motor- and nonmotor-related neurons. Neuroscience. 1995;67:313–324. [PubMed: 7675172]
  • Qiao JT, Dougherty PM, Wiggins RC, Dafny N. Effects of microiontophoretic application of cocaine, alone and with receptor antagonists, upon the neurons of the medial prefrontal cortex, nucleus accumbens and caudate nucleus of rats. Neuropharmacology. 1990;29:379–385. [PubMed: 2342637]
  • Ranck JB Jr., Kubie JL, Fox SE, Wolfson S, Muller RU. Single neuron recording in behaving mammal: bridging the gap between neuronal events and sensor-behavioral variables. In: Robinson T.E., editor. Behavioral approches to brain research. New York: Oxford University Press; 1983. Chap. 5.
  • Reddy L, Kanwisher N. Coding of visual objects in the ventral stream. Curr Opin Neurobiol. 2006;16:408–414. [PubMed: 16828279]
  • Richardson NR, Roberts DC. Progressive ratio schedules in drug self-administration studies in rats: a method to evaluate reinforcing efficacy. J Neurosci Methods. 1996;66:1–11. [PubMed: 8794935]
  • Robbins TW. The acquisition of responding with conditioned reinforcement: effects of pipradrol, methylphenidate, d-amphetamine, and nomifensine. Psychopharmacology (Berl) 1978;58:79–87. [PubMed: 27837]
  • Robbins TW, Everitt BJ. Functional studies of the central catecholamines. Int Rev Neurobiol. 1982;23:303–365. [PubMed: 6749738]
  • Robinson TE, Berridge KC. The neural basis of drug craving: an incentive-sensitization theory of addiction. Brain Res Brain Res Rev. 1993;18:247–291. [PubMed: 8401595]
  • Robinson TE, Berridge KC. Addiction. Annu Rev Psychol. 2003;54:25–53. [PubMed: 12185211]
  • Roesch MR, Stalnaker TA, Schoenbaum G. Associative encoding in anterior piriform cortex versus orbitofrontal cortex during odor discrimination and reversal learning. Cereb Cortex. 2007;17:643–652. [PMC free article: PMC2396586] [PubMed: 16699083]
  • Rolls ET, Thorpe SJ, Boytim M, Szabo I, Perrett DI. Responses of striatal neurons in the behaving monkey. 3. Effects of iontophoretically applied dopamine on normal responsiveness. Neuroscience. 1984;12:1201–1212. [PubMed: 6148716]
  • Salomone J. Report on Australian national conference: “Surrogacy—in whose interest?” Melbourne, February 1991. Issues Reprod Genet Eng. 1992;5:79–94. [PubMed: 11651336]
  • Sanger TD. Neural population codes. Curr Opin Neurobiol. 2003;13:238–249. [PubMed: 12744980]
  • Schmitzer-Torbert N, Jackson J, Henze D, Harris K, Redish AD. Quantitative measures of cluster quality for use in extracellular recordings. Neuroscience. 2005;131:1–11. [PubMed: 15680687]
  • Schultz W. Multiple reward signals in the brain. Nat Rev Neurosci. 2000;1:199–207. [PubMed: 11257908]
  • Schultz W, Apicella P, Ljungberg T. Responses of monkey dopamine neurons to reward and conditioned stimuli during successive steps of learning a delayed response task. J Neurosci. 1993;13:900–913. [PubMed: 8441015]
  • Schultz W, Apicella P, Scarnati E, Ljungberg T. Neuronal activity in monkey ventral striatum related to the expectation of reward. J Neurosci. 1992;12:4595–4610. [PubMed: 1464759]
  • Self DW, Choi KH, Simmons D, Walker JR, Smagula CS. Extinction training regulates neuroadaptive responses to withdrawal from chronic cocaine self-administration. Learn Mem. 2004;11:648–657. [PMC free article: PMC523085] [PubMed: 15466321]
  • Shaham Y, Shalev U, Lu L, De WitH, Stewart J. The reinstatement model of drug relapse: history, methodology and major findings. Psychopharmacology (Berl) 2003;168:3–20. [PubMed: 12402102]
  • Shalev U, Grimm JW, Shaham Y. Neurobiology of relapse to heroin and cocaine seeking: a review. Pharmacol Rev. 2002;54:1–42. [PubMed: 11870259]
  • Shidara M, Aigner TG, Richmond BJ. Neuronal signals in the monkey ventral striatum related to progress through a predictable series of trials. J Neurosci. 1998;18:2613–2625. [PubMed: 9502820]
  • Smith JE, Co C, Freeman ME, Lane JD. Brain neurotransmitter turnover correlated with morphine-seeking behavior of rats. Pharmacol Biochem Behav. 1982;16:509–519. [PubMed: 6123120]
  • Smith JE, Co C, Freeman ME, Sands MP, Lane JD. Neurotransmitter turnover in rat striatum is correlated with morphine self-administration. Nature. 1980;287:152–154. [PubMed: 6107854]
  • Spear DJ, Katz JL. Cocaine and food asreinforcers: effects of reinforcer magnitude and response requirement under second-order fixed-ratio and progressive-ratio schedules. J Exp Anal Behav. 1991;56:261–275. [PMC free article: PMC1323101] [PubMed: 1955816]
  • Stalnaker TA, Roesch MR, Franz TM, Calu DJ, Singh T, Schoenbaum G. Cocaine-induced decision-making deficits are mediated by miscoding in basolateral amygdala. Nat Neurosci. 2007;10:949–951. [PMC free article: PMC2562677] [PubMed: 17603478]
  • Stewart J, deWit H. Reinstatement of drug-taking behavior as a method of assessing incentive motivational properties of drugs. In: Bozarth M.A., editor. Methods of assessing the reinforcing properties of abused drugs. New York: Springer; 1987. pp. 211–228.
  • Stewart J, de Wit H, Eikelboom R. Role of unconditioned and conditioned drug effects in the self-administration of opiates and stimulants. Psychol Rev. 1984;91:251–268. [PubMed: 6571424]
  • Taha SA, Nicola SM, Fields HL. Cue-evoked encoding of movement planning and execution in the rat nucleus accumbens. J Physiol. 2007;584:801–818. [PMC free article: PMC2276984] [PubMed: 17761777]
  • Thomas MJ, Beurrier C, Bonci A, Malenka RC. Long-term depression in the nucleus accumbens: a neural correlate of behavioral sensitization to cocaine. Nat Neurosci. 2001;4:1217–1223. [PubMed: 11694884]
  • Thomsen M, Caine SB. Cocaine self-administration under fixed and progressive ratio schedules of reinforcement: comparison of C57BL/6J, 129Xl/SvJ, and 129S6/SvEvTac inbred mice. Psychopharmacology (Berl) 2006;184:145–154. [PubMed: 16369835]
  • Thomas MJ, Kalivas PW, Shaham Y. Neuroplasticity in the mesolimbic dopamine system and cocaine addiction. Br J Pharmacol. 2008;154(2):327–342. [PMC free article: PMC2442442] [PubMed: 18345022]
  • Tiffany ST. Cognitive concepts of craving. Alcohol Res Health. 1999;23:215–224. [PubMed: 10890817]
  • UchimuraN, North RA. Actions of cocaine on rat nucleus accumbens neurones in vitro. Br J Pharmacol. 1990;99:736–740. [PMC free article: PMC1917561] [PubMed: 2193689]
  • Uzwiak AJ, Guyette FX, West MO, Peoples LL. Neurons in accumbens subterritories of the rat: phasic firing time-locked within seconds of intravenous cocaine self-infusion. Brain Res. 1997;767:363–369. [PubMed: 9367270]
  • Vanderschuren LJ, Everitt BJ. Drug seeking becomes compulsive after prolonged cocaine self-administration. Science. 2004;305:1017–1019. [PubMed: 15310907]
  • Volkow ND, Wang GJ, Fowler JS, Hitzemann R, Angrist B, Gatley SJ, Logan J, Ding YS, Pappas N. Association of methylphenidate-induced craving with changes in right striato-orbitofrontal metabolism in cocaine abusers: implications in addiction. Am J Psychiatry. 1999;156:19–26. [PubMed: 9892293]
  • Wan X, Peoples LL. Amphetamine exposure enhances accumbal responses to reward-predictive stimuli in a pavlovian conditioned approach task. J Neurosci. 2008;28(30):7501–7512. [PubMed: 18650328]
  • Weddington WW, Brown BS, Haertzen CA, Cone EJ, Dax EM, Herning RI, Michaelson BS. Changes in mood, craving, and sleep during short-term abstinence reported by male cocaine addicts. A controlled, residential study. Arch Gen Psychiatry. 1990;47:861–868. [PubMed: 2393345]
  • West AR, Floresco SB, Charara A, Rosenkranz JA, Grace AA. Electrophysiological interactions between striatal glutamatergic and dopaminergic systems. Ann N Y Acad Sci. 2003;1003:53–74. [PubMed: 14684435]
  • West MO, Peoples LL, Michael AJ, Chapin JK, Woodward DJ. Low-dose amphetamine elevates movement-related firing of rat striatal neurons. Brain Res. 1997;745:331–335. [PubMed: 9037428]
  • White FJ, Henry D.J, Jeziorski M., Ackerman J.M. Electrophysiologic effects of cocaine within the mesoaccumbens and mesocortical dopamine systems. Boca Raton, FL: CRC Press; 1992.
  • White FJ, Hu XT, Henry DJ. Electrophysiological effects of cocaine in the rat nucleus accumbens: microiontophoretic studies. J Pharmacol Exp Ther. 1993;266:1075–1084. [PubMed: 8355182]
  • White FJ, Hu XT, Zhang XF. Neuroadaptations in nucleus accumbens neurons resulting from repeated cocaine administration. Adv Pharmacol. 1998;42:1006–1009. [PubMed: 9328068]
  • White FJ, Hu XT, Zhang XF, Wolf ME. Repeated administration of cocaine or amphetamine alters neuronal responses to glutamate in the mesoaccumbens dopamine system. J Pharmacol Exp Ther. (1995a);273:445–454. [PubMed: 7714800]
  • White SR, Harris GC, Imel KM, Wheaton MJ. Inhibitory effects of dopamine and methylenedioxymethamphetamine (MDMA) on glutamate-evoked firing of nucleus accumbens and caudate/putamen cells are enhanced following cocaine self-administration. Brain Res. (1995b);681:167–176. [PubMed: 7552276]
  • White FJ, Kalivas PW. Neuroadaptations involved in amphetamine and cocaine addiction. Drug Alcohol Depend. 1998;51:141–153. [PubMed: 9716936]
  • Whitelaw RB, Markou A, Robbins TW, Everitt BJ. Excitotoxic lesions of the baso-lateral amygdala impair the acquisition of cocaine-seeking behaviour under a second-order schedule of reinforcement. Psychopharmacology (Berl) 1996;127:213–224. [PubMed: 8912399]
  • Wise RA, Bozarth MA. A psychomotor stimulant theory of addiction. Psychol Rev. 1987;94:469–492. [PubMed: 3317472]
  • Wise RA, Newton P, Leeb K, Bumette B, Pocock D, Justice JB Jr. Fluctuations in nucleus accumbens dopamine concentration during intravenous cocaine self-administration in rats. Psychopharmacology (Berl) 1995;120:10–20. [PubMed: 7480530]
  • Wolf ME, Sun X, Mangiavacchi S, Chao SZ. Psychomotor stimulants and neuronal plasticity. Neuropharmacology. 2004;47 1:61–79. [PubMed: 15464126]
  • Woodward DJ, Chang J-Y, Janak P, Azarov A, Anstrom K. McGinty JF, editor. Mesolimbic neuronal activity across behavioral states. In: Advancing from the Ventral Striatum to the Extended Amygdala. New York: Ann. N.Y. Acad. Sci. 1999:91. [PubMed: 10415645]
  • Wyvell CL, Berridge KC. Intra-accumbens amphetamine increases the conditioned incentive salience of sucrose reward: enhancement of reward “wanting” without enhanced “liking” or response reinforcement. J Neurosci. 2000;20:8122–8130. [PubMed: 11050134]
  • Yoshimura M, Higashi H, Fujita S, Shimoji K. Selective depression of hippocampal inhibitory postsynaptic potentials and spontaneous firing by volatile anesthetics. Brain Res. 1985;340:363–368. [PubMed: 4027657]
  • Zhang GC, Mao LM, Liu XY, Wang JQ. Long-lasting up-regulation of orexin receptor type 2 protein levels in the rat nucleus accumbens after chronic cocaine administration. J Neurochem. 2007;103:400–407. [PubMed: 17623047]
  • Zhang XF, Hu XT, White FJ. Whole-cell plasticity in cocaine withdrawal: reduced sodium currents in nucleus accumbens neurons. J Neurosci. 1998;18:488–498. [PubMed: 9412525]
  • Zhang XF, Cooper DC, White FJ. Repeated cocaine treatment decreases whole-cell calcium current in rat nucleus accumbens neurons. Pharmacol Exp Ther. 2002;301:1119–1125. [PubMed: 12023545]

APPENDIX: DESCRIPTION OF TECHNIQUES

SURGERY

Intravenous Catheter

Catheter: A small Silastic tubing (0.025 o.d.) is inserted around the inferior part of the cannula, and then a larger Silastic tubing (0.046 o.d.) is inserted to reinforce insertion area around the cannula. A bubble of silicone is placed at 3.7 cm from the extremity of the small tubing. This extremity will be inserted into the right jugular vein. The opposite end of the catheter is connected to a 22-gauge guide cannula curved at 90 degrees in the inferior part. The 22-gauge cannula tubing is then cemented inside a nylon bolt. This terminal end of the catheter exits between the scapulae and is anchored there by means of sutures and a small piece of Marlex mesh.

Catheterization

Before the start of the surgery, rats are deeply anesthetized with ketamine (30 mg/ kg IP) and xylazine (5 mg/kg IP). The back and neck are clipped with electric clippers and then washed with betadine. A 1 cm incision is made throughout the skin of the neck diagonally from the mandibule to a point midway between midline and shoulder. Using forceps, the skin is separated from the muscle, and the muscle is bluntly dissected to expose the jugular vein. Using microforceps, the vein is isolated by teasing away surrounding muscle and tissue. A second incision is made in the back of the animal midway between the base of the tail and shoulders. The 22-gauge cannula tubing is placed under the skin, and the extremity of the cannula will exit in the middle of the back. Thereafter, the catheter is passed subcutaneously until it exits through the first incision, above the jugular vein. The jugular vein is cut, using Vannas scissors, and the catheter is inserted into the vein up to the bubble of silicone using a trocar. The catheter is secured to the vein with surgical silk sutures and then sutured to underlying smooth muscle. The silk suture is sterile and nonreactive and is necessary to provide a permanent anchoring of the catheter. The skin incisions are closed with individual chromic gut sutures, washed with sterile saline. Chromic gut suture is used to maximize rate of wound healing and minimize possibility of secondary injuries or infections associated with exposed suture accessible to the animal. A metal cap is placed on the extremity of the cannula to close the catheter. The wound is sprayed with gentamicin sulfate (GentavVed) topical spray.

Microwire Implant

Directly following the catheterization surgery, rats are placed into the stereotaxic apparatus and prepared for implant of the microwire array. The scalp incision is extended anterior to a point between the eyes. The facia is pushed back to the bone ridge, and the skull is thoroughly washed with saline. Using a dental drill, holes are drilled to accommodate six “skull screws.” We have traditionally use 1–72 × 1/8 slot pan m/s stainless steel screws from JET Fitting and Supply Corporation (Santa Ana, California). It can be helpful to use smaller screws in the most anterior plates of the skull. Screw holes should be placed fairly close to the midline of the skull. This allows adequate clearance between the screws and the bone ridge to apply dental cement. A rectangular hole, consistent with the rectangular shape of the array and slightly larger than the outer dimension of the array (hole dimension = 2.2 mm × 0.6 mm) is drilled through the skull above the accumbens (Paxinos and Watson 2004). An additional hole is drilled for the ground electrode. The skull is thoroughly cleaned with saline and allowed to dry.

The skull screws are then screwed into place. The ground wire is pushed under the skull and cemented to the skull screws. Cement is applied so as to completely cover screws and wires, but care is taken to avoid any cement interfering with the array implant. Thereafter, the connector strips of the array are attached to a custom-made “array holder” that is mounted on the stereotaxic apparatus. The array is then lowered into brain slowly and in stages to avoid dimpling of brain. The array is lowered in approximately 1-mm steps over 1-min periods. Each 1-mm lowering is separated by 5 min to allow the brain to recover. During the interval between successive lowering of the array, the well is filled with saline. This melts away polyethylene glycol from the array. During the lowering procedure care is taken to avoid contact between the wires and either the bone edges of the skull hole or any other implement that might scratch the wire. Before lowering the array the last 0.2 mm, the skull hole is filled with cement. Once the final lowering is made, the cement is allowed to thoroughly dry before continuing with the surgery. The array is then encased and cemented in place using dental acrylic. It should be noted that the dental acrylic does not bond to the Teflon wires; it is thus possible, given adequate pressure, to displace the microwires (e.g., pull them back up dorsally along the electrode track), at least at the early stages of applying the acrylic and before the connector is cemented into place. It is thus optimal to complete as much of the cementing before moving the connector end of the array.

Postoperative Care

Immediately after surgery, animals remain on a heating pad until they begin to move. Animals are then placed in the newly cleaned stainless steel grid holding cage. The catheters are flushed daily with 0.2 ml of an ampicillin solution (0.1 g/ml) containing heparin (300 IU/ml) to maintain patency, and the animals are checked to confirm that they heal well and that they are free of signs of pain and distress. We check for the following: normal consumption of food and water, evidence of normal elimination and grooming, normal movement around the home cage, and good wound condition. We allow animals a minimum of 7 days for recovery time before any further behavioral training.

RECORDING EQUIPMENT AND PROCEDURES

Microwire Array

Descriptions of microwires and various microwire arrays, along with photographs and diagrams, have been provided by other authors (e.g., Moxon 1999; Woodward et al. 1999). The array that we use is manufactured by Microprobes, Inc., (Gaithersburg, Maryland) and is viewable on their Web site (www.microprobes.com). Briefly, 16 microwires and 1 ground wire are soldered into two Omnetics (Minneapolis, Minnesota) miniature connector strips (GF-8). The solder connections are coated with epoxy, and the microwires are supported with polyethylene glycol to keep their spacing. Although the materials and configuration of the arrays can be customized, we use a rectangular array that consists of two rows of 8 quad-Teflon-coated stainless steel microwires (California Fine Wire, Grover City, California). The diameter of each wire in the array is 50 μm with no insulation. The two rows of microwires are separated from each other by 0.50 mm. Adjacent wires within each row are approximately 0.25 mm apart (wire center to wire center). This array configuration yields reliable success in recording the activity of single cells. Our early experience with microwire recording indicated that the use of the array had numerous advantages, relative to a bundle of wires. First, there is evidence that the yield of usable wires is greater and that the tissue damage is reduced by use of the array, relative to the bundle. Additionally, given the geometric configuration and between-wire spacing, it is possible to ultimately identify the location of each wire tip through histological analysis (discussed further below).

Recording Equipment

Each microwire in the array is connected via an independent channel to a computer-controlled recording system. We use a system built by Plexon, Inc. (Dallas, Texas), although comparable systems are made by other suppliers. All of these systems contain the same basic components, which include a head-mounted headstage amplifier, an electronic harness, a swivel, and a computer-controlled amplification and filtering system. These components will be described in detail in the following sections. All systems can be viewed in detail on the Web sites of the suppliers.

Tethering System

The tethering system contains the tether itself, a swivel, and a counterbalance. This system protects the catheter and electronic harness, transfers rotational force to the swivel, and provides strain relief for both the harness and the catheter.

The tether length is a critical variable that must be closely attended to. It influences the freedom of movement of the animal and the proper working of the tether system. A tether that is too short prevents the animal from moving to the outer edges and corners of the operant chamber. Because animals, particularly those exposed to drugs such as cocaine, move around a lot, the interface between the animal and the catheter or microwire array can be stressed. On the other hand, a tether that is too long tends to have inadequate torque and can twist on itself rather than rotate the swivel. This can lead to damage of the tether and thus the catheter or wire connections. A tether that is too long can also loop toward the animal during rearing and thus become accessible to the animal. Tethers are commercially available, although the length may need to be customized according to the dimensions of the operant chamber.

A swivel, or commutator, is necessary to reduce strain on the implant as the animal moves around. The swivel used in our experiment is the Airflyte CAY-675-24 electronic and fluid swivel (Airflyte Electronics, Bayonne, New Jersey). This swivel has low contact noise, low rotational torque, and O-ring seals that can withstand both alcohol sterilization and salt solutions. This swivel is additionally light and compact enough to be counterbalanced (see below). A simple fluid swivel is used for behavioral training when recordings are not in progress.

The counterbalance serves to reduce strain on the tether when the animal approaches the perimeter of the chamber while also moving the tether away from the animal when it rears. This serves to both reduce stress on the implant and catheter, and also reduce the possibility of the animal damaging the tether. Counterbalances are available commercially from Med Associates (St. Albans, Vermont), although they may need to be customized to accommodate the dimensions of the operant chamber.

OPERANT CHAMBERS

The operant chambers that we use are commercially purchased from Med Associates (St Albans, Vermont). The chamber is made largely of Plexiglas. The inner walls and floor of the chamber are free of any metal or protrusions. All devices (lights and speakers) are mounted on the outside of the chamber. The floors are removable and made of Plexiglas square rods. The operant chambers are housed in sound-attenuating outer chambers. In designing the placement of operant chambers and devices such as lights, pumps, and levers, it is important to anticipate situations that can introduce noise into the recordings. If the animal makes direct contact with a metallic part of the operant chamber, current may flow between the animal and that part, whch can cause disruptive artifacts in the recording. For this reason, we use nonconductive, plastic walls and coverings on all devices that the animal has access to. Devices such as retractable levers that are operated by DC motors can also generate electromagnetic fields that can cause artifacts without direct contact between the animal and the device. We have eliminated the potential for these artifacts by carefully grounding and shielding all devices that may generate such a field. The specific source of electrical artifacts will depend on the specifics of the experimental setup. Therefore, it is important to test for any potential sources of noise and to take steps to reduce the noise before carrying out experiments.

Recording Equipment

Voltage signals from individual microwires are first amplified 20× on a head-mounted headstage amplifier. Amplification at this early stage is important, because the signals are very small and would be degraded if they were transferred over large distances. The headstage amplifier is connected through the tether and swivel to a preamplifier, mounted outside of the operant chamber. This preamplifier amplifies the signal another 50× and also subtracts the signal of each microwire from a user-selected reference microwire. In general, the user selects a wire in the array that does not have any neural signals on it. Subtracting this reference signal from the signal on each microwire removes common sources of noise from all wires, such as noise caused by chewing or locomotion. This processed signal is transferred to a final amplification and filtering stage. The gain of this amplification is user selectable from 1× to 32×. This is useful for fine-tuning the gain of each channel to best detect neural signals. Also at this stage, a high-pass filter is applied to remove slow, cyclic activity such as the 60Hz noise. A user-selectable threshold is also applied to the neural signal at this stage, and all voltage events smaller than this threshold are discarded. Typically this threshold is set to be slightly larger than the noise band, such that large noise events and neural signals are retained. These events are digitized and stored for later offline analysis.

DATA ANALYSIS

After the experiment is over, the digitized waveforms are sorted into groups, so that action potentials from individual neurons can be separated from large noise events. Briefly, this procedure is called “spike-sorting” and consists of grouping the voltage events into groups of similar shapes, using parameters such as amplitude of the waveform, or principal components extracted from a sample set of waveforms. Three-dimensional scatterplots are used to plot these features against each other, which allows for visual discrimination of different groups as different clusters in the three-dimensional space. In general, there is a cluster of “noise” waveforms and one or more clusters of neural waveforms. Various controls are used to reduce the possibility that multiple neurons contributed to the activity in a cluster of neural waveforms. For example, accumbal neurons have a refractory period of ˜2msec. An automated procedure examines the interspike interval of all spikes and counts the percentage of spikes that occurred within 2 msec of each other. If this number is high, it may indicate that more than one neuron is contributing spikes to the cluster, considering it would be unlikely that a single neuron would fire so frequently within its own refractory period. It should be noted that these sorting procedures are far from standard, and the optimization and automation of them is the subject of ongoing research (Buzsaki 2004; Schmitzer-Torbert et al. 2005; Adamos et al. 2008; Chan et al. 2008). At the present time, commonly used methods of spike sorting involve a combination of computer-assisted and manual sorting of spikes. For additional discussion regarding data analysis, see Peoples (2003).

Histology

At the end of the experiment, histological procedures are used to confirm the location of the tip of the recording electrodes. Subjects are injected with a lethal dose of sodium pentobarbital. Anodal current (50 μA for 4 seconds) is passed through each microwire. Animals are perfused with formalin-saline. Coronal sections (50 μm) are mounted on slides and incubated in a solution of 5% postassium ferricyanide and 10% HC1 to stain the iron deposits left by the recording tip (Green 1958). The tissue is counterstained with 0.2% solution of Neutral Red. The location of each wire tip is plotted on the coronal plate (Paxinos and Watson 1998) that most closely corresponds to its anterior-posterior position. The microwire array is rectangular in configuration and made of two parallel rows of wires. The configuration and geometric relationship of the various wires to each other is usually maintained as the wires are lowered into brain so that it is possible to identify the location of each of the 12–16 wire tips during histology.

Copyright © 2010 by Taylor and Francis Group, LLC.
Bookshelf ID: NBK53363PMID: 21656978

Views

  • PubReader
  • Print View
  • Cite this Page

Other titles in this collection

Related information

  • PMC
    PubMed Central citations
  • PubMed
    Links to PubMed

Similar articles in PubMed

See reviews...See all...

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...