AUTHOR=Camarena Héctor O. , García-Leal Óscar , Delgadillo-Orozco Julieta , Barrón Erick 

TITLE=Probabilistic reinforcement precludes transitive inference: A preliminary study

JOURNAL=Frontiers in Psychology

VOLUME=Volume 14 - 2023

YEAR=2023

URL=https://www.frontiersin.org/journals/psychology/articles/10.3389/fpsyg.2023.1111597

DOI=10.3389/fpsyg.2023.1111597

ISSN=1664-1078

ABSTRACT=Transitive inference (TI) has been understood as a case of logical conclusion of the form if A>B>C>D>E, therefore B>D. This logical conclusion has been studied in several species, such as fishes, rodents, birds (e.g., pigeons and crows) and humans (e.g., children and adults). The basic procedure implies the training in a conditioned discrimination between pairs of stimuli: A+B-, B+C-, C+D-, D+E-. During the test the subject is exposed to untrained pairs (i.e., BD, AC, CE, AD, BE, AE). If the subject prefers B over D, the preference is regarded as evidence of TI. Several approaches have assumed that TI can be explained by differences in associative strength, instead of a logical conclusion. In the present experiment, probabilistic outcomes were introduced in a TI procedure to assess the effects of task complexity on TI. Our findings suggest that the introduction of probabilistic outcomes distorted the expected serial position effect (SPE), during training, and the symbolic distance effect (SDE), during test. Additionally, accuracy for the test pair BD was impaired. Thus, even tough, B preference was higher than D preference, the found preference did not differ from chance, however, some form of ordering of the stimuli was retained during test. The results are discussed in terms of discrimination deficits on learning and changes in associative strength provoked by probabilistic reinforcement.