“Causal reasoning” in rats: A reappraisal.

Abstract

It has recently been argued that rats engage in causal reasoning and they do so in a way that is consistent with Bayes net theories . This argument was based upon the finding that the tendency of cues to elicit approach to a food-well was reduced when their presentation was contingent on lever pressing. There is, however, an alternative interpretation of the critical experimental findings that is based on the simple principle of response competition: wherein lever pressing interferes with the tendency to approach the food well. Here the authors replicated Experiments 1 and 2a of and found reciprocal patterns of lever pressing and food well approach during the critical cues. These results lend direct support for an interpretation in terms of response competition while providing evidence contrary to Bayes net theories, and are readily interpreted within the theoretical framework provided by traditional associative learning theory. Keywords: Bayes casual nets, response competition, Lloyd MorganThe enduring debate within the field of comparative psychology concerns the degree to which the behavior of human and nonhuman animals is underpinned by the same cognitive mechanisms. Recent interest has focused on whether seemingly complex human behaviors can be understood in terms of simple (e.g., associative) processes that are ubiquitous in the animal kingdom, and whether nonhuman animals exhibit evidence of mechanisms considered to be uniquely human (e.g., causal reasoning). Recently, Blaisdell, Sawa, Leising, and Waldmann (2006a) have claimed that rats engage in causal reasoning-reasoning that is underpinned by the use of causal Bayes net models (see also Blaisdell, Sawa, Leising, & Waldmann, 2006b, 2007Leising, Wong, Waldmann, & Blaisdell, 2008). Evidence that is consistent with the suggestion that causal reasoning in humans is underpinned by causal Bayes nets comes from some simple observations that are analogous to those upon which Blaisdell et al. (2006a) made their claim.In humans who are aware that a single cause (e.g., raindrops) can have two separate effects (water droplets on window panes and dry clothes becoming damp on a washing line), the observation of one effect (wet windows) can lead to the inference that the other effect (damp clothes) has also occurred (this example is taken from Clayton & Dickinson, 2006). This inference presumably reflects the assumption that the observed effect (wet windows) must be the result of its usual cause (rain), which should also result in its second effect (damp clothes). However, this type of reasoning does not hold if there is an alternative cause (e.g., turning on a water sprinkler near to the window, but far away from the washing line) for the observed, first effect. The intervention, of turning on the sprinkler, provides an alternative cause for the first effect and the presence of the second effect is no longer inferred; and the response (e.g., removing the clothes from the line) is not required. The different patterns of inferences that obtain when an effect is simpl...

Associative foundation of causal learning in rats

“…Experiments 3 and 4 were conducted to differentiate between the response competition and causal accounts of the reduced leverpressing observed in Experiment 1. It should be noted that an account based on response competition here is different from that previously proposed (Dwyer et al, 2009) to account for prior claims of causal learning in rats (Blaisdell, Sawa, Leising, & Waldmann, 2006;Leising, Wong, Waldmann, & Blaisdell, 2008). In that case, two active responses (nose poking and leverpressing) competed with one another (see the General Discussion for further details).…”

“…The ability of nonhuman animals to make causal inferences about external stimuli has been a contested issue (e.g., Dwyer, Starns, & Honey, 2009). This series of experiments was designed to provide additional evidence concerning whether rats behave as if they have some sense of causal relations that we can examine within an operant sensory preconditioning experiment modeled after St. Claire-Smith and MacLaren (1983; see Table 1 for our design and the predictions based on the expectation of behavior suggestive of causal learning).…”

Intervening to learn about effects of one’s actions

“…Lever contact was greater during response-dependent CS presentations than response-independent CS presentations [F(1, 35.79) = 21.90; p < .001; cf. Dwyer et al, 2009], which may be regarded as a methodological artefact, since the former were initiated by lever contact, and lever contact occurs in spurts, such that lever contact predicts lever contact. Absence of an interaction of days with stimulus period [F(11, 92.36) = 0.94; p = .50] indicates that the increase of lever contact over days was similar during and prior to stimulus presentation.…”

“…It is a frequent assumption (e.g., Blaisdell et al, 2006;Dwyer et al, 2009;Fernbach & Sloman, 2009) that learners will take covariation information alone as evidence of direct causation. Seeing that a stimulus is often followed by food might lead one to interpret that stimulus as a cause of food.…”

“…Such theories have typically been developed as specifically about human learning, and their successes have so far not carried well to attempts to extend the perspective beyond humans to other animals (Bowers & Timberlake, 2017;Burgess et al, 2012;Dwyer, Starns, & Honey, 2009), despite the benefit of paths already blazed by researchers working with human subjects (e.g., Steyvers, Tenenbaumb, Wagenmakers, & Blum, 2003). One aim of the present study is to provide data to stimulate the development of theory that extends the use of graphical models to the learning of other species as well.…”