Rats maintain optimal choice when facing long terminal links in a “suboptimal choice” procedure.

Abstract

It has been recently reported that rats' optimal behavior in the "suboptimal choice" procedure is eliminated when the duration of the terminal links is increased to 30 s or more (Cunningham & Shahan, 2019). The main goal of the present study is to analyze the generality of such results, via a procedure that has been extensively used in the study of rats' performance in this task. In Experiment 1, nine rats were exposed to a procedure that presented levers as discriminative stimuli, varying the duration of their presentation from 10 s to 50 s across conditions. The results showed that rats preferred the optimal alternative in all conditions. To further analyze the relationship between rats' preferences and the duration of the terminal links, in Experiment 2 we contrasted the performance of two groups that differed in the configuration of the stimuli associated to the terminal links. For one of them, the stimuli were the same as in the Cunningham and Shahan's ( 2019) experiment (Tone þ Light); for the other, the stimulus was light only. The results show that rats were suboptimal in the former group, and optimal in the latter. We discuss our results in terms of the prevalence of rats' optimal behavior when confronted with the manipulation of different variables and the necessity of improving the knowledge about the relationship between optimal/suboptimal choice and the salience of the discriminative stimuli.

Suboptimal choice: A review and quantification of the signal for good news (SiGN) model.

“…Finally, although the current version of the SiGN model provides an excellent fit to a large set of choice results for pigeons and starlings without any free parameters, it seems likely that extensions or free parameters would be needed to fit the results of studies with rats or humans. For example, in studies with rats, the same temporal and probability parameters have yielded choice proportions that range from strong preference for the optimal alternative to strong preference for the suboptimal alternative depending on procedural variables such as the modalities or salience of the stimuli (e.g., Alba et al, 2021). Even within pigeons, use of a treadle press rather than a key peck as the choice response has produced different results (González-Torres et al, 2020).…”

“…They noted that rat studies showing suboptimal choice typically used a blackout as the S − and they found optimal choice when a distinct lever was presented as the S − . Alba et al (2021) failed to find suboptimal choice with long terminal links and simple stimuli, but they did find suboptimal choice at a long terminal link when the S + was a tone plus light, suggesting that salience of the discriminative stimuli is important. Across studies, our review suggests that under most conditions, rats are less likely than pigeons to choose the suboptimal alternative.…”

Differences in paradoxical choice between pigeons (Columba livia) and rats (Rattus norvegicus): The problem of cue trackability.

“…Of course, they are discriminable at the TL level because they are localized events (e.g., left vs. right on a wall). As noted earlier, the substantial work by Orduña and his colleagues demonstrated that the levers are discriminable, at the TL level, because they generate a number of responses that varies according to the probability of food that follows (e.g., Alba et al, 2021;López et al, 2018;Martínez et al, 2017). However, this property alone is possibly not sufficient to generate consistency tracking at the IL level.…”

“…Under the assumption that the evolutionary function of general search is to resolve uncertainty through consistency tracking (in opposition to focal search, which favors the approach to and interaction with a stimulus), the IL info alternative is the appropriate choice to achieve this goal. Accordingly, the IL stimuli are not much attributed to incentive salience, a process that does not appear to control choice behavior (Alba et al, 2021;Martínez et al, 2017;Trujano & Orduña, 2015). In other words, an IL info stimulus is chosen as a higher-level means allowing organisms to track the consistent TL Rats are less sensitive to the ratio between the delay to TL to the delay to food at the choice point than pigeons Vasconcelos and coll.…”

“…Such an interpretation requires further development through concrete examples. Daniels and Sanabria (2018) reported that paradoxical choice in rats seems to occur only if the TL+ and TL− stimuli are associated with distinct sensory modalities (see above, e.g., Alba et al, 2021;Chow et al, 2017;Cunningham & Shahan, 2019). However, the experimental results available do not demonstrate that two distinct sensory modalities are required; they only demonstrate that using sensorily too similar stimuli as TLs in the IL info alternative causes a preference for the IL noninfo alternative.…”