The role of temporal generalization in a temporal discrimination task

Asymmetry in updating long-term memory for time

The present study evaluates the updating of long-term memory for duration. After learning a temporal discrimination associating one lever with a standard duration (4 sec) and another lever with both a shorter (1-sec) and a longer (16-sec) duration, rats underwent a single session for learning a new standard duration. The temporal generalization gradient obtained 24 h later showed a modification in long-term memory for durations longer than the standard but only when the new duration was longer than the one initially learned. The effect was confirmed for another set of durations (0.5-2-8 sec). Our study demonstrates asymmetry in updating long-term memory for time.

Generalization of response patterns in a multiple peak procedure

Stimulus generalization is typically assessed by analyzing overall response rates. Studies of generalization of response-rate patterns across time are less common, despite the ubiquitous nature of time and the strong temporal control over behavior in the natural world. Thus, we investigated generalization of response-rate patterns across time using a multiple peak procedure in pigeons. The frequency (fast or slow) at which the color of a keylight changed signaled a fixed-interval (FI) 5-s or 20-s schedule, counterbalanced across subjects. In peak trials, the frequency of keylight-color changes was varied. For the fast and slow training stimuli, response rates in peak trials were controlled by the arranged FI schedule value; they increased as the arranged reinforcer time approached, and decreased thereafter. Response-rate patterns to all test stimuli were similar to response-rate patterns to the slow training stimulus for all subjects. Thus, overall, strong generalization from the slow training stimulus to all test stimuli was evident, whereas there was little to no generalization from the fast training stimulus. These findings extend past research examining generalization of temporally controlled response-rate patterns, and provide a useful starting point for future investigations of generalization of fixed-interval responding. A thorough understanding of generalization processes requires analysis of dependent variables other than overall response rates, especially when responding is likely to be temporally controlled.

Temporal generalization gradients following an interdimensional discrimination protocol

We investigated the effects of interdimensional discrimination training in the temporal generalization gradient. In a matching-to-sample task, pigeons learned to choose key S after a T-s houselight sample and key NS in the absence of the houselight sample. For one group of pigeons, T = 20 s; for another, T = 10 s. Subsequently, houselight duration was varied to obtain temporal generalization gradients. Results showed that (a) proportion S increased as houselight duration ranged from 0 s to T s and then remained high for houselight durations longer than T; (b) the gradients were well described by negative-exponential functions; (c) these non-flat gradients were present from the beginning of testing, and; (d) the average gradients obtained with T = 20 s and T = 10 s overlapped when plotted in relative time. We conclude that temporal control does not require explicit discrimination training along the temporal dimension, and that temporal generalization gradients obtained with an interdimensional protocol show the scalar property of timing. We discuss how these findings challenge current models of timing.

Relative versus absolute stimulus control in the temporal bisection task

When subjects learn to associate two sample durations with two comparison keys, do they learn to associate the keys with the short and long samples (relational hypothesis), or with the specific sample durations (absolute hypothesis)? We exposed 16 pigeons to an ABA design in which phases A and B corresponded to tasks using samples of 1 s and 4 s, or 4 s and 16 s. Across phases, we varied the mapping between the samples and the keys. For group Relative, short and long samples were always associated with the same keys (e.g., Phase A: '1s→ Left, 4s→ Right'; Phase B: '4s→ Left, 16s→ Right'); for group Absolute, the 4-s sample was associated always with the same key (e.g., Phase A: '1s→ Left, 4s→ Right'; Phase B: '16s→ Left, 4s→ Right'). If temporal control is relational, group Relative should learn the new task faster than group Absolute, but if temporal control is absolute, the opposite should occur. We compared the results with the predictions of the Learning-to-Time (LeT) model, which accounts for temporal discrimination in terms of absolute stimulus control and stimulus generalization. The acquisition curves of the two groups were generally consistent with LeT and therefore more consistent with the absolute than the relative hypothesis.

The interaction of temporal generalization gradients predicts the context effect

In a temporal double bisection task, animals learn two discriminations. In the presence of Red and Green keys, responses to Red are reinforced after 1-s samples and responses to Green are reinforced after 4-s samples; in the presence of Blue and Yellow keys, responses to Blue are reinforced after 4-s samples and responses to Yellow are reinforced after 16-s samples. Subsequently, given a choice between Green and Blue, the probability of choosing Green increases with the sample duration-the context effect. In the present study we asked whether this effect could be predicted from the stimulus generalization gradients induced by the two basic discriminations. Six pigeons learned to peck Green following 4-s samples (S(+)) but not following 1-s samples (S(-)) and to peck Red following 4-s samples (S(+)) but not following 16-s samples (S(-)). Temporal generalization gradients for Green and Red were then obtained. Finally, the pigeons were given a choice between Green and Red following sample durations ranging from 1 to 16 s. Results showed that a) the two generalization gradients had the minimum at the S(-) duration, an intermediate value between the S(-) and the S(+) durations, and the maximum at the S(+) as well as more extreme durations; b) on choice trials, preference for Green over Red increased with sample duration, the context effect; and c) the two generalization gradients predicted the average context effect well. The Learning-to-Time model accounts for the major trends in the data.

Learning-related shifts in generalization gradients for complex sounds

Learning to discriminate stimuli can alter how one distinguishes related stimuli. For instance, training an individual to differentiate between two stimuli along a single dimension can alter how that individual generalizes learned responses. In this study, we examined the persistence of shifts in generalization gradients after training with sounds. University students were trained to differentiate two sounds that varied along a complex acoustic dimension. The students were subsequently tested on their ability to recognize a sound that they had experienced during training when it was presented among several novel sounds varying along this same dimension. Peak shift was observed in Experiment 1, in which generalization tests immediately followed training, and in Experiment 2, in which the tests were delayed by 24 h. These findings further support the universality of generalization processes across species, modalities, and levels of stimulus complexity. They also raise new questions about the mechanisms underlying learning-related shifts in generalization gradients. The sound stimuli from this study are available as .wav files from http://lb.psychonomic-journals.org/content/supplemental.

Learning to Time: a perspective

In the last decades, researchers have proposed a large number of theoretical models of timing. These models make different assumptions concerning how animals learn to time events and how such learning is represented in memory. However, few studies have examined these different assumptions either empirically or conceptually. For knowledge to accumulate, variation in theoretical models must be accompanied by selection of models and model ideas. To that end, we review two timing models, Scalar Expectancy Theory (SET), the dominant model in the field, and the Learning-to-Time (LeT) model, one of the few models dealing explicitly with learning. In the first part of this article, we describe how each model works in prototypical concurrent and retrospective timing tasks, identify their structural similarities, and classify their differences concerning temporal learning and memory. In the second part, we review a series of studies that examined these differences and conclude that both the memory structure postulated by SET and the state dynamics postulated by LeT are probably incorrect. In the third part, we propose a hybrid model that may improve on its parents. The hybrid model accounts for the typical findings in fixed-interval schedules, the peak procedure, mixed fixed interval schedules, simple and double temporal bisection, and temporal generalization tasks. In the fourth and last part, we identify seven challenges that any timing model must meet.

Operant generalization in quail neonates after intradimensional training: Distinguishing positive and negative reinforcement

Operant generalization has been demonstrated in neonates only recently. To investigate the development of intradimensional stimulus control immediately after hatching, northern bobwhite chicks (Colinus virginianus) pecked for brief heat presentations while hearing a high-pitched sound repeated at two constant rates: an S+ tempo signaling a rich reinforcement schedule, alternating with an S- tempo signaling a leaner schedule. Tempo generalization was then assessed in extinction. The expected excitatory gradients were produced after a threshold number of training sessions; unexpectedly, below that threshold, gradients were inhibitory. The chicks' rapidly developing thermoregulatory capability may have resulted in a change from perceived negative reinforcement initially to positive reinforcement later. Given past research showing excitatory gradients after negative reinforcement, we suggest that these results demonstrate that all negative reinforcement is not equivalent, and, further, that classical conditioning effects require consideration.

Operant generalization of auditory tempo in quail neonates

Operant generalization, a building block of cognition, has never been studied in neonates. We developed a preparation for newly hatched quail in order to investigate the form of the generalization gradient very early in development. For their first 4-5 days after hatching, northern bobwhite chicks pecked for brief heat presentations while hearing a high-pitched sound repeated at a constant rate (nondifferential training). Tempo generalization was then assessed in extinction. A significantly excitatory average gradient resulted despite minimal postnatal experience.