Pigeons exhibit flexibility but not rule formation in dimensional learning, stimulus generalization, and task switching.

Abstract

A prominent model of categorization (Ashby, Alfonso-Reese, Turken, & Waldron, 1998) posits that 2 separate mechanisms—one declarative, one associative—can be recruited in category learning. These 2 systems can effectively be distinguished by 2 task structures: rule-based (RB) tasks are unidimensional and encourage analytic processing, whereas information-integration (II) tasks are bidimensional and encourage nonanalytic associative learning. Humans and nonhuman primates have been reported to learn RB tasks more quickly than II tasks; however, pigeons and rats have shown no learning speed differences are thus believed to lack the declarative system. In the present trio of experiments, we further explored pigeons’ dimensional category learning. We replicated the finding that pigeons learn RB and II tasks at equal speeds. Further, we found that stimulus generalization performance was equivalent on both tasks. We also explored the effect of switching from one task to another. Task switches between phases of training as well as within individual training sessions posed little difficulty for pigeons; they quickly and flexibly switched their categorization responses with no cost in choice speed or accuracy. Together, our data indicate that, although pigeons may lack the capacity to form explicit dimensional rules, their associative learning system is both powerful and flexible. Further exploration of this associative system would help us better appreciate possible contributions of the declarative system.

Prelimbic cortex maintains attention to category-relevant information and flexibly updates category representations

“…Category tasks were created by placing bivariate normal distributions on this transformed stimulus space ( Fig. 1 B; Category A: µ X = 30, σ X = 2.5, µ Y = 50, σ Y = 20; Category B: µ X = 70, σ X = 2.5, µ Y = 50, σ Y = 20; Broschard et al, 2019 , Broschard et al, 2020 , O’Donoghue et al, 2020 ). Each distribution constituted a category, and each point within a distribution represented a category stimulus.…”

“…5 A). Testing distributions were identical to the training distributions, except that the standard deviation along the relevant dimension (or axis for the 2D tasks) was increased ( σ X = 10; Broschard et al, 2019 , O’Donoghue et al, 2020 ). With this manipulation, some exemplars overlapped with the training distributions (i.e., Trained; within two standard deviations), but some exemplars sampled from novel portions of the stimulus space.…”

“…Reaction times from incorrect trials were excluded from all analyses. Additionally, reaction times that exceeded two standard deviations of the mean were excluded from all analyses, a criterion that is commonly used to eliminate outliers ( O’Donoghue et al, 2020 ). These outliers rarely occurred.…”

Pigeons proficiently switch among four tasks without cost.

“…Multiple species-including humans, pigeons, and macaqueshave proven to be highly adept at task switching (Avdagic et al, 2014;Caselli & Chelazzi, 2011;Meier et al, 2013;Meier et al, 2016b;O'Donoghue et al, 2020;Stoet & Snyder, 2003). However, different species may achieve their success through distinctly different means.…”

“…However, executive control need not be a necessary component of successful task switching. Of the various nonhuman species that have thus far been studied, some (e.g., macaques; Avdagic et al, 2014;Caselli & Chelazzi, 2011;Stoet & Snyder, 2003) are believed to possess at least some elements of human-like executive functioning, whereas others (e.g., pigeons; Meier et al, 2013;Meier et al, 2016b;O'Donoghue et al, 2020) are not. Yet, all of these species evidence a common behavioral hallmark: unlike humans, they rarely if ever exhibit switch costs (Caselli & Chelazzi, 2011, did report evidence of switch costs for rhesus macaques; however, that evidence may be an artifact of the researchers' statistical analysis; Avdagic et al.…”