|When Do Rats Wait for More Food? An Investigation of Reinforcer Accumulation
|Tuesday, May 31, 2016
|10:00 AM–11:50 AM
|Zurich FG, Swissotel
|Area: EAB; Domain: Basic Research
|Chair: Eric James French (Central Michigan University)
|Discussant: Eric A. Jacobs (Southern Illinois University Carbondale)
Reinforcer accumulation occurs when organisms forgo consuming available food in favor acquiring a cache of food to be consumed later. This symposium presents experiments that involve the factors that determine reinforcer accumulation or the effects of an obligate reinforcer accumulation contingency on responding. Bruner and Feregrino manipulated the temporal interval separating a component where lever pressing queued up reinforcers (procure component) that were delivered in a separate component (access component). Reinforcer accumulation increased as a function of the interval. Flores, Mateos, Madrigal and Bruner varied both the temporal interval separating procure and access components and the duration of the two components. They found that in addition to the temporal interval the duration of each component determined to the amount of reinforcers accumulated. French and Reilly investigated the independent effects inter-trial intervals (separating the access and procure component transitions) and response reinforcer delays (separating the procure and access component transitions) on reinforcer accumulation. Reinforcer accumulation increased when the response reinforcer delay was increased. Finally, Smith and Jacobs used a second-order token reinforcement schedule to investigate the effect of increasing the procure requirement on response rate when either two or four tokens had to be accumulated prior to the access component. Response rate was found to be a bitonic function of the procure requirement. The collective results of these studies have implications for response gradients, relative-time effects, delay discounting and conditioned reinforcement.
|Keyword(s): Reinforcer Accumulation, Reinforcer Delay, Tokens
|Food Accumulation as a Function of Delay of Reinforcement
|CARLOS A. BRUNER (National University of Mexico), Edith Feregrino (National University of Mexico)
|Abstract: In a previous investigation we studied food accumulation by rats. In each of 60 trials per session a 20-s access to each of two retractable levers was alternated. Each press to one lever (the procure lever) solely programmed an equal number of food pellets obtainable by pressing a second lever (the obtain lever). The interval between access to the procure and obtain levers was either 0, 1, 2, 4, 8, 16, 32 or 64 s on successive 30 session-blocks. We found that the number of accumulated food pellets increased as a function of lengthening the interval between access to the procure and the obtain levers. Using the same parameters of the previous study, in the present investigation the obtain lever was eliminated to avoid the possible induction from obtain to procure responses. After wait periods of either 0, 1, 4, 16 or 32 s one food pellet was delivered for each press on the 20-s access to the procure lever. As in the previous study pellet-accumulation increased as the wait period was lengthened. Aside from discarding response-induction as a possible artifact for food accumulation, these results challenge the established knowledge by showing that delay of reinforcement may result in increasing response gradients.
Access Time to Food-Procuring and Obtaining Levers Determines Food Accumulation by Rats
|CARLOS JAVIER FLORES AGUIRRE (Universidad de Guadalajara), Rebeca Mateos Morfán (Universidad de Guadalajara), Kenneth David Madrigal-Alcaraz (Universidad de Guadalajara)
Food accumulation in rats is an increasing function of lengthening the interval between access to two alternating retractable levers. Each press to the food-procuring lever programs the delivery of one pellet for each press on the food-obtaining lever. While in previous studies access to each lever has been held constant at 20 s, in the present investigation access to each lever was lengthened to either 40 or 60 s, with four rats in each condition. On successive phases the intervals between access to the levers were 0, 2, 8, and 32 s. For all rats the number of obtained food pellets increased in the range between 0 and 8 s. With 32 s the number of food pellets decreased. The number of food pellets and of procuring and obtaining responses was higher with lever-access constant at 40 s than at 60 s. It was concluded that lever-access duration controls both, the number of obtained pellets and the point of inflexion of the increasing food-accumulation functions, in a manner similar to relative-time effects.
|The Role of Inter-Trial Interval Duration and Response Reinforcer Delay on Reinforcer Accumulation
|ERIC JAMES FRENCH (Central Michigan University), Mark P. Reilly (Central Michigan University)
|Abstract: The purpose of the current experiment was to dissociate the effects that the inter-trial interval and response reinforcer delays have on reinforcer accumulation. In a within-subject design, rats made choices between two levers in a discrete trial procedure where each response on the left lever set up one reinforcer, to be collected later, and a single response on the right lever delivered the earned reinforcers. Conditions varied based on the presence of either an inter-trial interval (ITI) or a delay to reinforcement, and the duration of the temporal intervals (0, 5, 10, and 20-s). In general the delay to reinforcement condition produced greater accumulation than the ITI delay condition, and the effect was positively correlated with the duration of the temporal interval. In addition to increasing accumulation; time spent in the choice component increased across the delay to reinforcement values despite remaining relatively unchanged across the inter-trial interval conditions. These results challenge the utility of simple optimal-foraging models to account for the data while simultaneously expanding the generality of a temporal-discounting framework in explaining behavior temporally separated from reinforcement.
Second-Order Token-Production Schedule Performances Are (Mostly) Obedient to the Mathematical Principles of Reinforcement
|TRAVIS RAY SMITH (Southern Illinois University Carbondale), Eric A. Jacobs (Southern Illinois University Carbondale)
The present experiment investigated the effect of second-order token reinforcement schedules (requiring the accumulation of multiple tokens prior to receiving an opportunity to exchange the tokens for food) on response rates and evaluated those data in terms of Killeens (1994) Mathematical Principles of Reinforcement (MPR). Two Long-Evans Hooded rats (J164 & J165) were trained to earn and exchange steel ball-bearings (token reinforcers) for sweetened-condensed milk (unconditional reinforcers) throughout 2-hr sessions. The fixed-ratio (FR) token-production schedule requirement was increased following a block of three sessions (values varied between FR 2 FR 20). In condition 1, the exchange-production requirement was a second-order FR 2 schedule that required two tokens to be accumulated prior to transition to the token-exchange component. For condition 2, the exchange-production requirement was a second-order FR 4 schedule. The effects of successive token production across the second-order schedule components on response rates were evaluated and characterized using the MPR equation. The overall response rate data, plotted across the token-production schedule values, revealed the typical bitonic pattern that is described by the MPR equation. These data extend Killeens principles into second-order schedules and reveal some limitations of their application.