PMC

Columns A & B = Choice Situation (i.e., the unit of analysis which is defined by the area of daily life and type of choice being observed)
Columns C & D & E = Staff Offers A Choice? (i.e., the staff member’s behavior when choice is or is not offered to the resident)
Column F = Resident Accepts Offering Of Choice? (i.e., the resident’s behavior in response to the staff member’s initial offering of choice or in response to some other internal or external stimulus)
Columns G & H & I = Staff Enables The Choice Or Request? (i.e., the staff member’s behavior when the offered choice or the resident’s self-expressed request was or was not enabled)

Averted cases by age interval and county. In Allegheny County, relative to the No Choice strategy, the number of averted cases was 5494 in the Pediatric Choice, 15,581 in the Adult Choice, and 19,948 in the Choice for Both Age Groups strategies. In Wayne County, relative to the No Choice strategy, the number of averted cases was 9665 in the Pediatric Choice, 24,303 in the Adult Choice, and 34,054 in the Choice for Both Age Groups strategies. In Santa Clara County, relative to the No Choice strategy, the number of averted cases was 7819 in the Pediatric Choice, 25,245 in the Adult Choice, and 33,154 in the Choice for Both Age Groups strategies. In Sacramento County, relative to the No Choice strategy, the number of averted cases was 7069 in the Pediatric Choice, 19,445 in the Adult Choice, and 26,296 in the Choice for Both Age Groups strategies. In Salt Lake County, relative to the No Choice strategy, the number of averted cases was 5597 in the Pediatric Choice, 14,316 in the Adult Choice, and 20,040 in the choice for both strategies.

Panel A shows the timing of daily choice and extended-access sessions. During choice sessions (11am–1pm), subjects could choose between food and increasing heroin doses. During extended-access sessions (1pm-10am), food was not available, but subjects could respond for 0.1 mg/kg/injection heroin. During time-out periods, neither food nor heroin was available. Panel B shows heroin choice dose-effect curves when heroin was available only during daily choice sessions (“Choice Alone”), during both choice sessions and extended-access sessions (“Choice + Extended Access”), or on the first day after termination of extended-access sessions (“Choice + Withdrawal Day 1). Abscissa: Unit heroin dose in mg/kg/injection, log scale. Ordinate: Percent of total choices allocated to heroin choice during each component of the choice session. Panel C shows effects of morphine treatment on withdrawal-associated increases in heroin choice over the entire choice session. Abscissa: Treatment conditions during extended heroin access (Ext Acc), after 1 day of withdrawal from extended heroin access (WD), or after 1 day of withdrawal during which subjects were also treated with continuous infusions of morphine (dose in mg/kg/hr). Ordinate: Percent of total choices allocated to heroin choice for the entire choice session. Asterisks indicate significantly different from WD (p<0.05). All points show mean ± SEM from three (B) or four (C) monkeys. Data adapted from ; .

Conventions are the same as in (A–D) Predictions of the CDA model. The overall value of an option in different locations of the attribute space is plotted for different choice set size: (from left to right) zero or in the absence of context effects, two, four, and eight options. The option value monotonically increases with the set size for any locations of the attribute space. (B) The average decoy value (over the attribute space) as a function of the choice set size in the presence and absence of the context effects. For each point we average the decoy value over all locations of the attribute space in panel B. (C) The average of the sum of option values in the choice set as a function of the choice set size. Conventions are the same as in B. The total options values exponentially increases with the choice set size. (D) Required computational resources for the network counterpart of the CDA model. Conventions are the same as in B. The computational resources required for context effects also exponentially grow with the choice set size. (E–H) Predictions of the RN model. In the RN model, the range of possible values of an option does not depend on the choice set size. (F) The average decoy value as a function of the choice set size. Conventions are the same as in B. The average decoy value does not depend on the choice set size. (G) The average of the sum of option values in the choice set as a function of the choice set size. The overall value of options in the choice set only increases linearly with the choice set size. Note the difference in scale in G and C. (H) Required computational resources for the RN model. In contrast to the CDA model, computational resources required by the RN model only increases linearly with the choice set size.