Total number of nose touches emitted by shelter dogs in each session under Food or Social (SOC) conditions (and Extinction—EXT—in the case of Dog S) (a-i). A total of 15 nose touches was possible in each session. Dogs that experienced the Social condition first are graphed on the left; dogs that experienced the Food condition first are graphed on the right. Graph j shows the medians (height of bar) and interquartile ranges (whiskers) of the total number of nose touches per session emitted by all dogs in the last complete Food (dark gray) and Social (diagonal hatching) conditions for each dog (i.e., three sessions of each condition).

Medians (height of bar) and interquartile ranges (whiskers) of the latencies to respond for shelter dogs in Experiment 1. The data graphed are the latencies from the last complete Food (dark gray) and Social (diagonal hatching) conditions for each dog (i.e., three sessions of each condition). The maximum latency for any trial was 60 s.

Cumulative records for all sessions for 2 individual shelter dogs from Experiment 2, both corresponding to the most common pattern of responding our study revealed. Other details are the same as for Figure 2. Graph a shows the most common pattern of responding (short latencies in the Food condition and long latencies in the Social condition). Graph b shows the next most common pattern of responding (low and variable responding in both conditions).

Total number of nose touches emitted by owned dogs in each session under Food, Social (SOC), and Extinction (EXT) conditions (a-f). Dogs that experienced the Social condition first are graphed on the left; dogs that experienced the Food condition first are graphed on the right. Graph g shows the medians and interquartile ranges of the total number of nose touches per session emitted by all owned dogs. Other details are the same as for Figure 4g.

Medians and interquartile ranges of the latencies to respond for owned dogs. Details are the same as for Figure 6.

Total number of nose touches emitted by hand-reared, captive wolves in each session under Food, Social (SOC), and Extinction (EXT) conditions (a-i). Wolves that experienced the Social condition first are graphed on the left; wolves that experienced the Food condition first are graphed on the right. Graph j shows the medians and interquartile ranges of the total number of nose touches per session emitted by all wolves. Other details are the same as for Figure 4g.

Medians and interquartile ranges of the latencies to respond for hand-reared, captive wolves. Details are the same as for Figure 6.

Cumulative records for all sessions for 3 individual shelter dogs from Experiment 1, each corresponding to one of three patterns of responding our study revealed. Time during reinforcer delivery was subtracted, leaving a cumulative record of time from stimulus onset to response. The dashed vertical lines show when conditions changed (condition type noted at the top of each graph). Arrows show where each individual session ended. Asterisks at the top of the graph indicate when a session ended before the dog emitted all 15 possible nose touches (i.e., the dog stopped responding for at least 1 min). Graph a shows the most common pattern of responding (short latencies in the Food condition and long latencies in the Social condition). Graph b shows the next most common pattern of responding (low and variable responding in both conditions). Graph c shows the pattern of responding for Dog S, who was unique in our study for responding equivalently to food (FD) and social interaction (SOC) consequences.

Total number of nose touches emitted by shelter dogs in each session under Food, Social (SOC), and Extinction (EXT) conditions (a-f). Dogs that experienced the Social condition first are graphed on the left; dogs that experienced the Food condition first are graphed on the right. Graph g shows the medians and interquartile ranges of the total number of nose touches per session emitted by all shelter dogs in the last complete Food (dark gray), Social (diagonal hatching), and Extinction (white) conditions for each dog (i.e., three sessions of each condition).

Medians and interquartile ranges of the latencies to respond for shelter dogs in Experiment 2. The data graphed are the latencies from the last complete Food (dark gray), Social (diagonal hatching), and Extinction (white) conditions for each dog (i.e., three sessions of each condition). Other details are the same as for Figure 3.

Cumulative records for all sessions for 2 individual owned dogs from Experiment 3, both corresponding to the most common pattern of responding our study revealed. One dog experienced the Social condition first (a) and 1 dog experienced the Food condition first (b). Other details are the same as for Figure 5.

Total number of nose touches emitted by shelter dogs in each session under extended Social conditions, followed by a Food condition.

Cumulative records for all sessions for 2 individual wolves from Experiment 5, both corresponding to the most common pattern of responding. One wolf experienced the Social condition first (a) and the other experienced the Food condition first (b). Other details are the same as for Figure 5.