The sequential view: From rapidly fading stimulus traces to the organization of memory and the abstract concept of number

Abstract

The development ofthe sequential approach to instrumentallearning from about 1958 to the present is described. The sequential model began as an attempt to explain a particular dass of neglected partial reward phenomena, those in which performance in acquisition and extinction is influenced by the particular sequence in which rewarded and nonrewarded trials occur in acquisition, and it was subsequently applied to a variety of other phenomena. Over time, the sequential model grew, sometimes through the replacement of older assumptions by novel ones, as when retrieved memories replaced stimulus traces, and sometimes simply through the addition of novel assumptions, such as that animals are capable of remembering retrospectively one, two, three or more prior nonrewarded outcomes--the N-Iength assumption. The most recent assumption added to the sequential model is that on a given trial the animal may utilize its memory of prior reward outcomes to anticipate both the current reward outcome and one or more subsequent reward outcomes. One way to view the sequential model is to say that it is a specific theory in various degrees of competition with other specific theories. Several examples of this are provided. Another way to view the sequential model, a more important way in my opinion, is to see it as a representative of a general theoretical approach, intertrial theory, which differs in fundamental respects from another much more generaIly utilized theoretical approach, intratrial theory. I suggest that there is a substantial body of data that can be explained by intertrial mechanisms but not by intratrial mechanisms. The future may weIl reveal that the intertrial mechanisms have greater explanatory potential than the currently more popular intratrial mechanisms.The cognitive abilities of rats are far greater and more varied than was dreamed of even remotely until recently. Even now they are familiar to only a relative few. As Wright (1992) has recently indicated, an animal may perform poorly not because of its cognitive limitations, but because the experimenter has employed a task ill suited to the particular species. Varied reward situations, emphasized in this paper, have demonstrated themselves to be ideally suited to revealing a variety of complex abilities in rats. To be sure, varied reward situations were employed initiaIly to test the simplest of hypotheses. But varied reward situations, as we shall see, are very flexible, and when appropriately modified they can be ernployed in the service of many ends. Within the context ofthe author's sequential hypothesis, varied reward tasks have been repeatedly modified to reveal increasingly complex processes in the rat, an activity that is ongoing.In the present paper, varied reward refers to situations with the following characteristics. Animals receive two or more different reward outcomes, such as food reward Correspondence concerning this article should besent to E. J. Capaldi, Department of Psychological Sciences, Purdue University, 1364 Psychology Building, West ...

Slower reacquisition after partial extinction in human contingency learning.

“…Later, when the reinforced trials appear during reacquisition, their presence should help activate not only the excitatory context of acquisition but also the inhibitory association formed during extinction, causing a slower reacquisition than in a standard extinction condition where only the acquisition context is retrieved. Consistently with Capaldi's (1967Capaldi's ( , 1994 sequential learning theory, Ricker and Bouton (1996) suggested that reacquisition is controlled by a "trial-signaling" mechanism, whereby the animal learns that a certain type of trial, either a reinforced or a non-reinforced trial, reliably signals the type of upcoming trial. As a result, during acquisition animals not only learn that the cue is paired by the reinforcer, but also that reinforced trials follow other reinforced trials, whereas during standard extinction animals learnt that non-reinforced trials follow other non-reinforced trials.…”

Forming chunks in instrumental learning: The role of overshadowing

“…As a concrete example, in Hull's (1943) the:' ory, the strength of the tendency to make such intratrial responses was said to increase when they were reinforced and to decrease when they were nonreinforced. The intratrial conception of the response, despite its general popularity, is difficult to reconcile with a wide variety ofempirical findings (see, e.g., Capaldi, 1992Capaldi, , 1994Capaldi, Miller, Alptekin, & Barry, 1990).…”

Retention of Ordinal Position Information with Limited and Extended Serial Training

“…For example, an animal may have received the series SNS' and PNS', or PNP' and SNP' (where S' and P' are twice the magnitude of Sand P). In these series, the memory of N could be strongly associated with an upcoming reward, and the memory of reward could be associated with N (Capaldi, 1994). Ordinal position could also be associated with reward outcomes because the first and third positions are always rewarded and the second position is never rewarded (Burns et aI., 1999).…”

“…Though position learning is now emphasized (e.g., Henson, 1998) in human cognition, it began receiving attention (e.g., Bower, 1971 ;Ebenholtz, 1972) long after memoryassociation theories (e. g., Ebbinghaus, 1913) had developed rich traditions. The study of memories, usually of differing reward events, in animal serial learning also has a long history that has produced clear evidence for animals employing reward memories in serial problems; and the understanding of the nature of those memories is now fairly sophisticated (reviewed by Capaldi, 1994). Evidence that animals use information about the position of items in a list is recent, but also clear (Burns, Dunkman, & Detloff, 1999;Burns, Kinney, & Criddle, (2000) ; Burns, Wiley, & Payne, 1986;Chen, Swartz, & Terrace, 1997;Couvillon, Brandon, Woodard, & Bitterman, 1980; D 'Amato & Colombo, 1988, Roiblat, Pologue, & Scopatz, 1983Straub & Terrace, 1981).…”

“…In this procedure, the well-established pattern of running fast on the first and third trials and slow on the second trial was retained in all three kinds of series in the transfer tests, a result that shows the influence of learning about position information , but tells us little about the nature of that information. Burns et al (2000) offered several possible sources of information about position that included: intentional or unintentional trial-to-trial variations in running procedures (Burns et aI., 1986), the time (e.g., Church & Broadbent, 1991) from the first trial to each of the remaining trials in the series, the memory of reward outcomes from previous trials (Capaldi, 1994), and the number of previous trials in the series (e.g., Capaldi & Miller, 1988). It may also be that position fashions a linear, cognitive representation (D ' Amato, 1991), which the animal may scan from position to position.…”