Conditioned place preferences in humans using secondary reinforcers.

The goal of this experiment was to examine whether a conditioned place preference could be established in humans using a secondary reinforcer that provided little obvious reward to the participants. Two experiments were conducted to answer this question. In Experiment 1, 244 undergraduates were placed into a VR environment consisting of two visually distinct rooms connected by a door. Throughout the experiment, one room was randomly paired with occasional point rewards while the other unique room was never paired with rewards. Participants received thee pairings in each room. After a short break, a test session was administered, and participants were given free access to the entire VR environment and no point rewards were administered. On the test day, we observe that participants displayed a significant CPP for the room paired with points, as evidenced by significant differences in rating each of the rooms in terms of enjoyment. In Experiment 2, 77 undergraduates were tested using a biased conditioning approach in which an initial test session was conducted to obtain the participant's preferred room bias, and then the least-preferred room was designated as the points reward room for each participant. Using this biased conditioning approach, participants spent a significantly greater amount of time in the points-paired room. In this case, participants showed preferences based on explicit and implicit measures. These results suggest new approaches to examine the role of secondary reinforcers in nontraditional addictions such as internet, gaming, and gambling dependencies.

Pavlovian conditioning to food reward as a function of eating disorder risk.

The aim of this experiment was to examine the extent to which eating disorder risk affects the strength of food-reward conditioning. Eighty food-restricted undergraduates were placed into a VR environment consisting of two visually distinct rooms. Participants underwent multiple pairing sessions in which they were confined into one of the two rooms and explored a VR environment. Room A was paired with real-life M&Ms for three sessions, and Room B was paired with no food for three sessions. After a short delay, a test session was administered, and participants were given free access to the entire VR environment for 5 min. Participants also completed the Eating Attitudes Test (EAT-26; [11]), which is a standard screening tool of eating disorder risk. Participants displayed a significant conditioned place preference for the VR room previously paired with food, and they displayed a significant explicit preference for the M&M-paired room in a forced-choice test. There was a significant positive correlation between place preference strength and scores on the dieting subscale of the EAT-26. Additionally, ratings of the no-food room were significantly lower as dieting scores increased. This suggests that components of eating disorder risk can influence basic conditioning strength to places associated with food reward. For both males and females, additional correlations between eating disorder risk subscales and conditioning variables are discussed, and implications for future research are proposed in hopes of understanding how conditioning paradigms can provide insight into treating and preventing eating disorders.

Conditioned place preferences in humans using virtual reality.

To extend a standard paradigm of conditioning in nonhumans to humans, we created a virtual reality (VR) conditioned place preference task, with real-life food rewards. Undergraduates were placed into a VR environment consisting of 2 visually distinct rooms. On Day 1, participants underwent 6 pairing sessions in which they were confined into one of the two rooms and explored the VR environment. Room A was paired with real-life M&Ms for 3 sessions, and Room B was paired with no food for 3 sessions. Day 2 was the test day, administered the next day, and participants were given free access to the entire VR environment for 5min. In experiment 1, participants were food restricted, and we observed that on the test day, participants display a significant conditioned place preference for the VR room previously paired with food (p<0.001). Additionally, they display a significant explicit preference for the M&M-paired room in a forced-choice of "Which room do you like best?". In experiment 2, when participants were not food restricted, there was no evidence of a place preference, either implicitly (e.g. dwell time) or explicitly. Hence, we show that we can reliably establish a place preference in humans, but that the preference is contingent on the participants' hunger state. Future research will examine the extent to which these preferences can be blocked or extinguished as well as whether these preferences are evident using other reinforcers.