Optimality And Concurrent Variable-interval Variable-ratio Schedules

Despite claims to the contrary, all leading theories about operant choice may be seen as models of optimality. Although melioration is often contrasted with global maximization, both make the same core assumptions as other versions of optimality theory, including momentary maximizing, hill climbing, and the various versions of optimal foraging theory. The present experiment aimed to test melioration against more global optimality and to apply the visit-by-visit analysis suggested by foraging theory. Rats were exposed to concurrent schedules in which one alternative was always variable-ratio 10 and the other alternative was a variable-interval schedule. Although choice relations varied from rat to rat, the overall results roughly confirmed the matching law, a result often taken to support melioration. Pooling the data across sessions and across rats, however, resulted in no increment in unsystematic variance, lending support to the contention by Ziriax and Silberberg (1984) that the choice relation is partly constrained. When the data were analyzed at the level of visits, the results either disconfirmed predictions of melioration or showed regularities about which melioration is silent. Instead, performance tended toward a rough optimization, in which responding favored the variable ratio, but with relatively brief visits to the variable interval. There were no asymmetries in travel or variability that would indicate that different processes were involved in generating visits at the two different schedules. The findings point toward a more global optimality model than melioration and demonstrate the value of per-visit analysis in the study of concurrent performances.     

Learning by Collaborative and Individual‐Based Recommendation Agents

Intelligent recommendation systems can be based on 2 basic principles: collaborative filters and individual‐based agents. In this work we examine the learning function that results from these 2 general types of learning‐smart agents. There has been significant work on the predictive properties of each type, but no work has examined the patterns in their learning from feedback over repeated trials. Using simulations, we create clusters of “consumers” with heterogeneous utility functions and errorful reservation utility thresholds. The consumers go shopping with one of the designated smart agents, receive recommendations from the agents, and purchase products they like and reject ones they do not. Based on the purchase–no purchase behavior of the consumers, agents learn about the consumers and potentially improve the quality of their recommendations. We characterize learning curves by modified exponential functions with an intercept for percentage of recommendations accepted at Trial 0, an asymptotic rate of recommendation acceptance, and a rate at which learning moves from intercept to asymptote. We compare the learning of a baseline random recommendation agent, an individual‐based logistic‐regression agent, and two types of collaborative filters that rely on K‐mean clustering (popular in most commercial applications) and nearest‐neighbor algorithms. Compared to the collaborative filtering agents, the individual agent (a) learns more slowly, initially, but performs better in the long run when the environment is stable; (b) is less negatively affected by permanent changes in the consumer's utility function; and (c) is less adversely affected by error in the reservation threshold to which consumers compare a recommended product's utility. The K‐mean agent reaches a lower asymptote but approaches it faster, reflecting a surprising stickiness of target classifications after feedback from recommendations made under initial (incorrect) hypotheses.     

DYNAMICS OF CHOICE: RELATIVE RATE AND AMOUNT AFFECT LOCAL PREFERENCE AT THREE DIFFERENT TIME SCALES

To examine extended control over local choice, the present study investigated preference in transition as food‐rate ratio provided by two levers changed across seven components within daily sessions, and food‐amount ratio changed across phases. Phase 1 arranged a food‐amount ratio of 4:1 (i.e., the left lever delivered four pellets and the right lever one pellet); Phase 2 reversed the food‐amount ratio to 1:4, and in Phase 3 the food‐amount ratio was 3:2. At a relatively extended time scale, preference was described well by a linear relation between log response ratio and log rate ratio (the generalized matching law). A small amount of carryover occurred from one rate ratio to the next but disappeared after four food deliveries. Estimates of sensitivity to food‐amount ratio were around 1.0 and were independent of rate ratio. Analysis across food deliveries within rate‐ratio components showed that the effect of a small amount was diminished by the presence of a large amount—that is, when a larger amount was present in the situation (three or four pellets), the value of a small amount (one or two pellets) became paltry. More local analysis of visits to the levers between food deliveries showed that postfood visits following a large amount were disproportionately longer than following a small amount. Continuing food deliveries from the same source tended to make visits less dependent on relative amount, but a discontinuation (i.e., food from the other lever) reinstated dependence on relative amount. Analysis at a still smaller time scale revealed preference pulses following food deliveries that confirmed the tendency toward dependence on absolute amount with continuing deliveries, and toward dependence on relative amount following discontinuations. A mathematical model based on a linear‐operator equation accounts for many of the results. The larger and longer preference following a switch to a larger amount is consistent with the idea that local preference depends on relatively extended variables even on short time scales.     

Memoria y adquisición del conocimiento

RESUMEN

En este artículo se sitúan los trabajos clásicos de Craik y Lockhart y Craik y Tulving en el contexto del estudio global de la memoria. Se destaca el énfasis de los dos últimos en los procesos en detrimento de las descripciones estructurales, situando la clave en el nivel de profundidad al que se procese la información y subrayando la importancia de relacionarla con nuestro conocimiento del mundo. Con relación al estudio de Craik y Tulving se examinan tres cuestiones fundamentales: el procedimiento de aprendizaje incidental, el concepto de profundidad de procesamiento y la noción de memoria episódica. Para ello revisan algunos estudios que cuestionan que el procesamiento tenga lugar únicamente de abajo a arriba a través de una jerarquía de estadios fijos -según propusieron Craik y Lockhart-, abordando diversos intentos de especificar el concepto de nivel de profundidad. La última parte se reserva al problema de identificar la noción de significado psicológico con la noción de significado procedente de la Lingüística, presentando las aportaciones de los teóricos “neogibsonianos”. El trabajo finaliza con algunas reflexiones respecto al concepto de memoria episódica aportado por Tulving.     

Reconnaissance en traduction et homographie interlangue

Several studies on bilingual word recognition have shown effects of word similarity between languages. Cognate words (translation equivalents with identical or near-identical forms like LIBRE in French and Spanish) are generally recognized and translated faster than non-cognates (translation equivalents with different forms). In this study, a translation recognition task (de Groot, 1992) was used in which participants (French-Spanish bilinguals) had to decide whether two words presented on a computer screen were translations or not. In Experiment 1, translation equivalents were identical cognates (same form: CIVIL-CIVIL [civil in Spanish]) and non-cognates (different forms: DANSE-BAILE [dance in Spanish]). All non-translation equivalents had different forms (TABLE [table in French]-AMIGO [friend in Spanish]). We observed a facilitation effect for cognate pairs which were processed faster than non-cognate pairs. In Experiment 2, we used the same material for translation equivalents (cognates and non-cognates) and two types of non-translation equivalents: interlingual homographs (same form but different meanings: CREER [create in French]-CREER [believe in Spanish]) and non-homographic non-translation pairs (different forms between languages) as used in Experiment 1. When the non-translation pairs shared the same form (interlingual homographs), they were rejected more slowly than other non-translation pairs. Moreover, contrary to Experiment 1, due to the presence of interlingual homographs in the experimental lists, the facilitation effect for cognate pairs was not replicated. The results suggest that all homographs (cognates and interlingual homographs) have a special status in bilingual memory (due to their lexical and/or semantic overlap) but their processing also depends on task demands and experimental list composition. Our results are in line with the distributed conceptual feature model of bilingual memory ( [de Groot, 1992] and [van Hell et de Groot, 1998] ). This model can explain facilitation and inhibition effects due to different overlaps between words (in both lexical and semantic levels). However, our results lead us to distinguish identification processes and decisional processes in this task as described in the BIA+ model (Dijkstra and van Heuven, 2002).