Enhancing "self-control": The paradoxical effect of delay of reinforcement

Autoshaped impulsivity: Some explorations with a neural network model

This study evaluated the effect of delay and magnitude of reinforcement in Pavlovian contingencies, extending the understanding of the phenomenon of autoshaped impulsivity as described in Alcalá's thesis (2017) and Burgos and García-Leal (2015). The effects of adding a trace interval were analyzed on the maintained responses of impulsive choice, seen as the preference of a small and immediate reinforcer over a larger and delayed one, and the role of the contextual unit, as well as the inhibitory units according to the Diffuse Discrepancy Model. In the Simulation, the model with inhibitory units was used, trained in two signals with different delays and reinforcement magnitudes, and subsequently presented concurrently in choice tasks without reinforcement nor learning, using an ABA within-subject design. In general, the DD model successfully simulated the phenomenon of autoshaped impulsivity, consistent with studies from Alcalá's thesis (2017), Burgos and García-Leal (2015), and Picker and Poling (1982). It also predicted the elimination of this effect (autoshaped impulsivity) after introducing a trace interval. The observed results and their implications are discussed, as well as possible future studies with animals and humans.

"What you see may not be what you get": Reverse contingency and perceived loss aversion in pigeons

The reverse-contingency task is a task in which one is given a choice between two rewards, but one receives the larger amount only if one chooses the smaller amount. This task is very difficult for chimpanzees unless the choice is between symbolic representations of the amounts. We found that pigeons can learn this task easily, if the reward amounts are associated with distinctive colors and the choice is delayed by 5 s. The reverse-contingency task involves three components: a loss when choosing one alternative, a gain when choosing the other, and the contrast between what was expected and what occurred. In Experiment 2 we separated the loss from the gain and found that experiencing a loss is sufficient for pigeons to learn to avoid that alternative. Finally, we found evidence for perceived loss aversion. When pigeons were offered a small amount of food and they received that amount, they preferred it over an alternative that offered them a larger amount but gave them only the smaller amount (a perceived loss). The results indicate that loss aversion, based on reference dependence, is likely a general phenomenon, and not only found in humans and other primates. We suggest that it can be attributed to contrast, the difference between what is expected and what is obtained, and it is related to the endowment effect and the mere ownership effect found in humans.