Behavior stability and individual differences in Pavlovian extended conditioning

Abstract

How stable and general is behavior once reached maximum learning? To answer this question and understand post-acquisition behavior and its related individual differences, we propose a psychological principle that naturally extends the basic associative single-cue Rescorla-Wagner model (which may also be called Hull model) of Pavlovian conditioning to (i) a framework of dynamical models predicting resistance to learning in the first few sessions followed by an over-optimal response peak. In turn, (ii) the theory can be further extended to describe response fluctuations by the laws of quantum mechanics. (iii) We also introduce an independent model characterized by the presence of a stochastic noise of cognitive origin. We ran an unusually long experiment with 32 rats over 3960 trials, where we excluded habituation and other well-known phenomena as sources of variability in the subjects' performance. There is weak to positive evidence that Hull model is the best nonlinear regression to averaged data only for a minority of the subjects, while its dynamical extension (i) can explain the totality of data. The noise encountered in all individual responses is white, thus confirming the simplest version of model (iii) but standing in contrast with the colored-noise findings in human performance. Finally, data are compatible with the quantum extension (ii), although experimental uncertainties dominate the result. On the other hand, data do not favor models with (iv) a long-memory effect or (v) where response variability is solely described by a random fractal.

Beyond Rescorla-Wagner: the ups and downs of learning

“…We will find up to very strong evidence that the DOM can fit more individual data than the RW model and the NAOM. This analysis will significantly extend and confirm the findings of Calcagni et al (2020) with more data, more statistics, and with the additional, robust cross-check represented by the NAOM.…”

“…In parallel, we will also explore the nature of the random, trial-by-trial response fluctuations characterizing the data. On the one hand, a spectral analysis will confirm that these fluctuations are described by white noise (Calcagni et al, 2020). On the other hand, in Appendix C we will further explore an associative but nondeterministic model constructed in the same paper with the tools of quantum mechanics, where random response variations are a quantitative part of the theory.…”

“…We are talking about the principle of least action, well known in physics but recently adapted to the psychological core of Pavlovian conditioning (Calcagni et al, 2020).…”

“…Example apparent to the naked eye are Figure 6 in Ghirlanda and Ibadullaiev (2015) (where, if not oscillations, at least an overshoot is visible), Figure 3 in Miller, Greco, and Vigorito (1981), and, notably, Figure 2 in Zelikowsky and Fanselow (2010). These are not the only available examples and, in fact, oscillations (not to be confused with the shortest-scale phenomenon of post-peak depression; for a discussion, see Calcagni, Caballero-Garrido and Pellón, 2020) may be more common than currently acknowledged, partly because there are not looked for or are ignored as experimental errors. In turn, they might not be looked for because, to begin with, the existing theories do not predict their occurrence.…”

“…In this article, we use data from different experiments to test one such candidate, proposed in Calcagni et al (2020): A Pavlovian associative dynamical oscillatory model (henceforth DOM) where the subject's response fluctuates periodically above and below the ideal asymptote of learning until these long-range oscillations (spanning several tens of trials) eventually disappear. 1 Using the Bayes and Akaike Information Criteria in a standard model selection procedure, we reanalyze data presented in Harris et al (2015) and Calcagni et al (2020) and compare the DOM with the RW model as well as with a non-associative oscillating model (NAOM) constructed as an ad hoc modification of the RW model designed to mimic response oscillations. We will find up to very strong evidence that the DOM can fit more individual data than the RW model and the NAOM.…”