Academic journal article The Psychological Record

Delay Gradients for Spout-Licking and Magazine-Entering Induced by a Periodic Food Schedule

Academic journal article The Psychological Record

Delay Gradients for Spout-Licking and Magazine-Entering Induced by a Periodic Food Schedule

Article excerpt

Most research on delay discounting studies the tradeoff between delays to rein forcers and expected rein forcer magnitudes. Because the same kind of response is used (e.g., different lever-presses, key pecks, or verbal mands for each alternative), the issue of special relationships between types of response and delayed rein forcers does not arise. However, the literature suggests that some responses are better able to sustain delays to reinforcement than others. In his singular chapter on self-control, Skinner (1953) noted that one part of the human repertoire may be used to control another--we may find it easier to bite our tongue or slap our hand over our mouth than to otherwise stifle a comment that might be offensive; or to pinch ourselves or deprive or satiate ourselves to change emotional and motivational states, else the behaviors that they foster have long-term negative consequences.

Chelonis and Logue (1996) compared the behavior of pigeons pecking keys or pressing treadles for different delays of reinforcement in a concurrent-chain paradigm. The average slope of the regressions in logarithmic coordinates (the sensitivity) was less for treadle pressing (0.24 vs. 0.32), but the substantial variability among pigeons kept this difference from statistical significance. In a more recent study comparing key pecking and treadle pressing, Holt and associates (Holt et al., 2013) found that pigeons demonstrated steeper discounting in the key peck condition than in the treadle press condition. In another study, Chelonis and associates (Chelonis, Logue, Sheehy, & Mao, 1998) showed that as the force required to operate their levers increased, rats made increasing numbers of self-control choices in the standard small-soon large-late (SSLL--Ainslie, 1974; Rachlin & Green, 1972) paradigm.

Differences in type of response can thus affect self-control. Differences in the magnitude or quality of the reinforcer can also: Chelonis and Logue (1997) found that rats' "self-control" for water, measured as choice of the larger, delayed outcome, was less than that for food. Human delay gradients for hypothetical outcomes are often shallower for larger magnitudes of reinforcement (Green, Myerson, & Ostaszweski, 1999). No magnitude effect has been reliably shown in nonverbal animals (Green, Myerson, Holt, Slevin, & Estle, 2004; Ong & White, 2004).

Schedules of Differential Reinforcement

Whereas the Smaller Sooner, Larger Later (SSLL) choice paradigm is excellent for assessing the tradeoff between the utility of different magnitudes of reinforcement at different delays, it tells us more about preference structures than it does about self-control or delay gradients (Killeen, 2015a). When we humans choose a SS outcome, we seldom call our choice impulsive, but rather defend it as a simple preference. A better measure of self-control may derive from differential reinforcement schedules in which low or zero responding is required to obtain reinforcement or avoid punishment. Such schedules have some ecological validity: The failure to avoid temptations--to not stay in seat, not talk out of turn not fidget, not cheat--is diagnostic for some impulse control disorders (Killeen, Tannock, & Sagvolden, 2012). Omission contingencies, such as differential reinforcement of zero response rates (DR0), differential reinforcement of other behavior (DRO, "other" because of the assumption that a rein forcer always reinforces something when not the target response), or differential reinforcement of alternative behavior (DRA, if the competing response is explicitly reinforced), have not appeared as often in the literature on impulsivity as has the choice paradigm. In fact, these "DR-" schedules are a type of choice paradigm, as the animal must choose between making the target response soon, a time when there is only a small probability of its delivering reinforcement, or later, waiting until the response will have a higher probability of success. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.