In many non-human species, learning retention decreases temporarily following training. This has led to the suggestion that these lapses reflect a fundamental component of memory formation. If so, transient memory lapses should also be prevalent in humans, and should occur for all types of learning. In line with these predictions, we report two cases of transient memory lapses in humans that occur 1-3 h after training on a perceptual-discrimination task. The results indicate that the occurrence of transient memory lapses extends to perceptual learning, a form of skill learning, and suggest that transient memory lapses may be a common but overlooked facet of memory formation in humans.

Musicians perform better than non-musicians on a variety of non-musical sound-perception tasks. Whether that musicians' advantage extends to spatial hearing is a topic of increasing interest. Here we investigated one facet of that topic by assessing musicians' and non-musicians' sensitivity to the two primary cues to sound-source location on the horizontal plane: interaural-level-differences (ILDs) and interaural-time-differences (ITDs). Specifically, we measured discrimination thresholds for ILDs at 4 kHz (n =246) and ITDs at 0.5 kHz (n = 137) in participants whose musical-training histories covered a wide range of lengths, onsets, and offsets. For ILD discrimination, when only musical-training length was considered in the analysis, no musicians' advantage was apparent. However, when thresholds were compared between subgroups of non-musicians (<2 years of training) and extreme musicians (≥10 years of training, started ≤ age 7, still playing) a musicians' advantage emerged. Threshold comparisons between the extreme musicians and other subgroups of highly trained musicians (≥10 years of training) further indicated that the advantage required both starting young and continuing to play. In addition, the advantage was larger in males than in females, by some measures, and was not evident in an assessment of learning. For ITD discrimination, in contrast to ILD discrimination, parallel analyses revealed no apparent musicians' advantage. The results suggest that musicianship is associated with greater sensitivity to ILDs, a fundamental sound-localization cue, even though that sensitivity is not central to music, that this musicians' advantage arises, at least in part, from nurture, and that it is governed by a neural substrate where ILDs are processed separately from, and more malleably than, ITDs.

Perceptual learning is the process by which experience alters how incoming sensory information is processed by the brain to give rise to behavior—it is critical for how humans educate children, train experts, treat diseases, and promote health and well-being throughout the lifespan. Knowledge of perceptual learning requires basic and applied research in humans and nonhuman animal models, which informs strategic targets for advancing applications. Commercial products to induce perceptual learning are proliferating rapidly with limited regulation (e.g., for rehabilitation), while at the same time basic science is increasingly restricted by changing regulations (such as new granting-agency definitions of clinical trials). Realizing the full potential of perceptual learning requires balancing basic and translational science to advance new knowledge, while serving and protecting consumers. Reforms can promote open, accessible, and representative research, and the translation of this research to applications across different sectors of society.

Training on one task (task A) can disrupt learning on a subsequently trained task (task B), illustrating anterograde learning interference. We asked whether the induction of anterograde learning interference depends on the learning stage that task A has reached when the training on task B begins. To do so, we drew on previous observations in perceptual learning in which completing all training on one task before beginning training on another task (blocked training) yielded markedly different learning outcomes than alternating training between the same two tasks for the same total number of trials (interleaved training). Those blocked versus interleaved contrasts suggest that there is a transition between two differentially vulnerable learning stages that is related to the number of consecutive training trials on each task, with interleaved training presumably tapping acquisition, and blocked training tapping consolidation. Here, we used the blocked versus interleaved paradigm in auditory perceptual learning in a case in which blocked training generated anterograde—but not its converse, retrograde—learning interference (A→B, not B←A). We report that anterograde learning interference of training on task A (interaural time difference discrimination) on learning on task B (interaural level difference discrimination) occurred with blocked training and diminished with interleaved training, with faster rates of interleaving leading to less interference. This pattern held for across-day, within-session, and offline learning. Thus, anterograde learning interference only occurred when the number of consecutive training trials on task A surpassed some critical value, consistent with other recent evidence that anterograde learning interference only arises when learning on task A has entered the consolidation stage.

Dave Green had an enormous influence on auditory research via signal-detection theory. A major contribution of signal-detection theory is the concept and application of the ideal detector, which establishes the absolute optimal detection performance. To bridge between the ideal performance and the less-than-ideal human performance, Green and Swets (1966) described an energy detector. The energy-detector model relaxed assumptions of the ideal detector to account for the phase insensitivity and limited frequency selectivity of humans, but it retained the assumption that the time window is matched to the signal duration. However, the time-window assumption had no clear support from human data. To help fill this gap, we examined the detectability of signals with expected versus unexpected temporal properties. The results of experiments and signal-detection analyses suggest that listeners listen selectively to signal duration, implying that they match the time window to the overall signal duration as was assumed in the energy-detector model. Additional results suggest that listeners do not also listen selectively to the temporal structure of the signal, implying that the temporal constraint is based on overall signal duration. We are grateful for the opportunity to have worked with Dave.

The two primary cues to sound-source location on the horizontal plane are interaural differences in time (ITDs) and in level (ILDs). Here, we asked whether the ability to discriminate small changes in each of these interaural cues differs between the sexes. We tested males and females who had no prior experience with any psychoacoustic task on either ILD discrimination at 4 kHz or ITD discrimination at 0.5 kHz. For ILD discrimination, the overall mean threshold, as well as the threshold for each 50-trial block, was significantly lower for males (n = 80) than for females (n = 166). Both males and females learned over the course of testing, but there was no sex difference in learning rate. In contrast, for ITD discrimination, thresholds did not differ significantly between the sexes for the overall mean or for any block (n = 43M/94F). There also was no learning across blocks for either group. For both tasks, the individual thresholds spanned a wide range in each group. The presence of sex differences and learning for ILD but not ITD discrimination suggests that the factors responsible for these outcomes influenced a relatively peripheral ILD-specific pathway, and not an ITD-specific pathway, or a more central locus that is not cue specific.

Distinguishing between regular and irregular heartbeats, conversing with speakers of different accents, and tuning a guitar-all rely on some form of auditory learning. What drives these experience-dependent changes? A growing body of evidence suggests an important role for non-sensory influences, including reward, task engagement, and social or linguistic context. This review is a collection of contributions that highlight how these non-sensory factors shape auditory plasticity and learning at the molecular, physiological, and behavioral level. We begin by presenting evidence that reward signals from the dopaminergic midbrain act on cortico-subcortical networks to shape sound-evoked responses of auditory cortical neurons, facilitate auditory category learning, and modulate the long-term storage of new words and their meanings. We then discuss the role of task engagement in auditory perceptual learning and suggest that plasticity in top-down cortical networks mediates learning-related improvements in auditory cortical and perceptual sensitivity. Finally, we present data that illustrates how social experience impacts sound-evoked activity in the auditory midbrain and forebrain and how the linguistic environment rapidly shapes speech perception. These findings, which are derived from both human and animal models, suggest that non-sensory influences are important regulators of auditory learning and plasticity and are often implemented by shared neural substrates. Application of these principles could improve clinical training strategies and inform the development of treatments that enhance auditory learning in individuals with communication disorders.

Most sounds fluctuate in amplitude, but do listeners attend to the temporal structure of those fluctuations when trying to detect the mere presence of those sounds? This question was addressed by leading listeners to expect a faint sound with a fixed temporal structure (pulse train or steady-state tone) and total duration (300 ms) and measuring their ability to detect equally faint sounds of unexpected temporal structure (pulse train when expecting steady state) and/or total duration (<300 ms). Detection was poorer for sounds with unexpected than with expected total durations, replicating previous outcomes, but was uninfluenced by the temporal structure of the expected sound. The results disagree with computational predictions of the multiple-look model, which posits that listeners attend to both the total duration and temporal structure of the signal, but agree with predictions of the matched-window energy-detector model, which posits that listeners attend to the total duration but not the temporal structure of the signal. Moreover, the matched-window energy-detector model could also account for previous results, including some that were originally interpreted as supporting the multiple-look model. Taken together, at least when detecting faint sounds, listeners appear to attend to the total duration of expected sounds but to ignore their detailed temporal structure.

Extended high frequencies (EHF), above 8 kHz, represent a region of the human hearing spectrum that is generally ignored by clinicians and researchers alike. This article is a compilation of contributions that, together, make the case for an essential role of EHF in both normal hearing and auditory dysfunction. We start with the fundamentals of biological and acoustic determinism e humans have EHF hearing for a purpose, for example, the detection of prey, predators, and mates. EHF hearing may also provide a boost to speech perception in challenging conditions and its loss, conversely, might help explain difficulty with the same task. However, it could be that EHF are a marker for damage in the conventional frequency region that is more related to speech perception difficulties. Measurement of EHF hearing in concert with otoacoustic emissions could provide an early warning of age-related hearing loss. In early life, when EHF hearing sensitivity is optimal, we can use it for enhanced phonetic identification during language learning, but we are also susceptible to diseases that can prematurely damage it. EHF audiometry techniques and standardization are reviewed, providing evidence that they are reliable to measure and provide important information for early detection, monitoring and possible prevention of hearing loss in populations at-risk. To better understand the full contribution of EHF to human hearing, clinicians and researchers can contribute by including its measurement, along with measures of speech in noise and self-report of hearing difficulties and tinnitus in clinical evaluations and studies. (C) 2020 Published by Elsevier B.V.

Semi-supervised learning refers to learning that occurs when feedback about performance is provided on only a subset of training trials. Algorithms for semi-supervised learning are popular in machine learning because of their minimal reliance on labeled data. There have been, however, only a few reports of semi-supervised learning in humans. Here we document human semi-supervised learning on a nonnative phonetic classification task. Classification performance remained unchanged when 60 feedback trials were provided on each of the two days of training. In contrast, performance improved when 60 feedback trials were combined with 240 no-feedback trials each day. In variants of this successful semi-supervised regimen, increasing the daily number of feedback trials from 60 to 240 did not increase the amount of learning, while decreasing that number to 30 abolished learning. Finally, replacing the no-feedback trials with stimulus exposure alone had little effect on the outcome. These results were an unexpected consequence of combining training periods with feedback and testing periods without feedback, illustrating that no-feedback testing can influence learning outcomes. More broadly, these data suggest that task performance with feedback can function as an all-or-none trigger for recruiting the contribution of trials without feedback, or mere stimulus exposures, to human learning.

What is required for long-lasting learning of a non-native phonetic contrast? Do principles from fine-grained discrimination learning apply? Discrimination learning often requires extensive practice within a day for performance to improve across days, and a 30-minute break midway through practice can disrupt that learning. Thus it appears that trials must integrate to a learning threshold in a transient memory store in order for durable discrimination learning to occur. We asked whether this same principle applies in phonetic-contrast learning. To do so, we trained monolingual speakers on a non-native phonetic contrast along a voice-onset-time (VOT) continuum. Those who engaged in a single continuous practice session on day 1 reached above chance accuracy on day 2 while classifying the non-native phoneme. In contrast, when day-1 practice included a 30-minute break halfway through training, day-2 performance was at chance. These results suggest that for phoneme learning, as for discrimination learning, long-lasting learning requires that practice trials be integrated up to a learning threshold within a transient memory store before they are sent en masse into a memory that lasts across days.

Auditory skills improve with practice. Producing across-day learning requires a sufficient number of trials per session; additional trials in the same session produce no additional improvement. Thus, trials appear to integrate within a session to reach a learning threshold, after which there is a refractory period during which trials do not contribute to learning. The refractory period duration must be <24 hours because training on consecutive days yields learning each day, but its exact duration is unknown. To narrow the range of possible durations, we trained young-adult, normal-hearing listeners on a basic auditory task (interaural-level-difference discrimination) either for two sessions of sufficient training separated by 30 minutes (n = 10) or 10 hours (n = 10), or for just one session (n = 15), and assessed their learning on the next day. The 30-minute group learned no more than the single-session group, indicating the second session did not aid learning, whereas the 10-hour group learned more than the other two groups, indicating the second session provided additional benefit. These results suggest that the refractory period lasts >30 minutes, but <10 hours, and does not require sleep to reset. Knowledge of the refractory period duration will improve the efficiency of perceptual training for typical and clinical populations.

For perceptual learning on fine-grained discrimination tasks, improvement can be enhanced or disrupted when two tasks are trained, depending on how the training on those tasks is distributed. To investigate if this phenomenon extends to speech learning, we trained native-English speakers to transcribe both Mandarin- and Turkish-accented English sentences using one of three different configurations of the same training stimuli. After training, all trained groups performed better than untrained controls, and similarly to each other, on a novel talker of a trained accent (Mandarin). However, for a novel accent (Slovakian), performance was better than untrained controls when training alternated between the two accents, but not when the two accents were trained consecutively. Performance for the novel accent decreased as the number of contiguous sentences per accent during training increased. One interpretation of these results is that accent information is integrated during a restricted time window. If two accents are encountered within this window, information from both accents is integrated, yielding accent-general learning. If two accents are encountered in separate consecutive windows, accent-specific learning occurs for each accent and accent-general learning is prevented. These results mirror patterns for fine-grained discrimination learning, and illuminate the processes underlying the success of high-variability training.

A key component of musical proficiency is the ability to discriminate between and identify musical intervals, or fixed ratios between pitches. Acquiring these skills requires training, but little is known about how to best arrange the trials within a training session. To address this issue, learning on a musical-interval comparison task was evaluated for two four-day training regimens that employed equal numbers of stimulus presentations per day. A regimen of continuous practice yielded no learning, but a regimen that combined practice and stimulus exposure alone generated clear improvement. Learning in the practice-plus-exposure regimen was due to the combination of the two experiences, because two control groups who received only either the practice or the exposure from that regimen did not learn. Posttest performance suggested that this improvement in comparison learning generalized to an untrained stimulus type and an untrained musical-interval identification task. Naïve comparison performance, but not learning, was better for larger pitch-ratio differences and for individuals with more musical experience. The reported benefits of the practice-plus-exposure regimen mirror the outcomes for fine-grained discrimination and speech tasks, suggesting that a general learning principle is involved. In practical terms, it appears that combining practice and stimulus exposure alone is a particularly effective configuration for improving musical-interval perception.

Perceptual learning is an enhancement in discriminability of similar stimuli following experience with those stimuli. Here, we examined the efficacy of adding additional active training following a standard training session, compared with additional stimulus exposure in the absence of associated task performance. Mice were trained daily in an odor-discrimination task, and then, several hours later each day, received 1 of 3 different manipulations: 1) a second active-training session, 2) non-task-related odor exposure in the home cage, or 3) no second session. For home-cage exposure, odorants were presented in small tubes that mice could sniff and investigate for a similar period of time as in the active discrimination task each day. The results demonstrate that daily home-cage exposure was equivalent to active odor training in supporting improved odor discrimination. Daily home-cage exposure to odorants that did not match those used in the active task did not improve learning, yielding outcomes similar to those obtained with no second session. Piriform cortical local field potential recordings revealed that both sampling in the active learning task and investigation in the home cage evoked similar beta band oscillatory activity. Together the results suggest that odor-discrimination learning can be significantly enhanced by addition of odor exposure outside of the active training task, potentially because of the robust activity evoked in the olfactory system by both exposure paradigms. They further suggest that odorant exposure alone could enhance or maintain odor-discrimination abilities in conditions associated with olfactory impairment, such as aging or dementia.

Rhythm is fundamental to music and speech, yet little is known about how even simple rhythmic patterns are processed. Here we investigated the processing of isochronous rhythms in the short inter-onset-interval (IOI) range (IOIs < 250–400 ms) using a perceptual-learning paradigm. Trained listeners (n=8) practiced anisochrony detection with a 100-ms IOI marked by 1-kHz tones, 720 trials per day for 7 days. Between pre- and post-training tests, trained listeners improved significantly more than controls (no training; n=8) on the anisochrony-detection condition that the trained listeners practiced. However, the learning on anisochrony detection did not generalize to temporal-interval discrimination with the trained IOI (100 ms) and marker frequency (1 kHz) or to anisochrony detection with an untrained marker frequency (4 kHz or variable frequency vs. 1 kHz), and generalized negatively to anisochrony detection with an untrained IOI (200 ms vs. 100 ms). Further, pre-training thresholds were correlated among nearly all of the conditions with the same IOI (100-ms IOIs), but not between conditions with different IOIs (100-ms vs. 200-ms IOIs). Thus, it appears that some task-, IOI-, and frequency-specific processes are involved in fast-rhythm processing. These outcomes are most consistent with a holistic rhythm-processing model in which a holistic “image” of the stimulus is compared to a stimulus-specific template.

Some forms of associative learning require only a single experience to create a lasting memory [1, 2]. In contrast, perceptual learning often requires extensive practice within a day for performance to improve across days [3, 4]. This suggests that the requisite practice for durable perceptual learning is integrated throughout each day. If the total amount of daily practice is the only important variable, then a practice break within a day should not disrupt across-day improvement. To test this idea, we trained human listeners on an auditory frequency-discrimination task over multiple days and compared the performance of those who engaged in a single continuous practice session each day [4] with those who were given a 30-min break halfway through each practice session. Continuous practice yielded significant perceptual learning [4]. In contrast, practice with a rest break led to no improvement, indicating that the integration process had decayed within 30 min. In a separate experiment, a 30-min practice break also disrupted durable learning on a non-native phonetic classification task. These results suggest that practice trials are integrated up to a learning threshold within a transient memory store before they are sent en masse into a memory that lasts across days. Thus, the oft cited benefits of distributed over massed training [5, 6] may arise from different mechanisms depending on whether the breaks occur before or after a learning threshold has been reached. Trial integration could serve as an early gatekeeper to plasticity, helping to ensure that longer-lasting changes are only made when deemed worthwhile.

Many perceptual abilities differ between the sexes. Because these sex differences have been documented almost exclusively in adults, they have been attributed to sex-specific neural circuitry that emerges during development and is maintained in the mature perceptual system. To investigate whether behavioral sex differences in perception can also have other origins, we compared performance between males and females ranging in age from 8 to 30 years on auditory temporal-interval discrimination and tone-in-noise detection tasks on which there are no sex differences in adults. If sex differences in perception arise only from the establishment and subsequent maintenance of sex-specific neural circuitry, there should be no sex differences during development on these tasks. In contrast, sex differences emerged in adolescence but resolved by adulthood on two of the six conditions, with signs of a similar pattern on a third condition. In each case, males reached mature performance earlier than females, resulting in a sex difference in the interim. These results suggest that sex differences in perception may arise from differences in the maturational timing of common circuitry used by both sexes. They also imply that sex differences in perceptual abilities may be more prevalent than previously thought based on adult data alone.

Detection of a tonal signal in amplitude-modulated noise can improve with increases in noise bandwidth if the pattern of amplitude fluctuations is uniform across frequency, a phenomenon termed comodulation masking release (CMR). Most explanations for CMR rely on an assumption that listeners monitor frequency channels both at and remote from the signal frequency in conditions that yield the effect. To test this assumption, detectability was assessed for signals presented at expected and unexpected frequencies in wideband amplitude-modulated noise. Detection performance was high even for signals of unexpected frequency, suggesting that listeners were monitoring multiple frequency channels, as has been assumed.

Listeners learn non-native phonetic contrasts given practice and feedback, but what is the dose-response curve for that training? Does testing without feedback affect that response? We trained native English speakers on a non-native phonetic contrast between pre-voicing and voicing on two consecutive days, varying the number of daily training trials across groups. Practice with feedback for 240, 120, or 60 trials/day all led to a similar steepening of the category-boundary slope, but practice with feedback for 30 trials/day yielded no learning. To determine whether the daily pre- and post-training tests without feedback (120 trials/test) affected this outcome, we trained another group 60 trials/day with feedback, but removed all but the final post-test. This group did not learn, indicating the tests without feedback facilitated improvement. Finally, to establish whether the benefit from these tests came from task performance, we trained another group 60 trials/day with feedback, but replaced the original tests with stimulus exposure alone until the final post-test. This group’s ending performance resembled that obtained with task performance during the tests without feedback. Thus, trials with feedback can function as an all-or-none trigger for recruiting the contribution of trials without feedback, or mere stimulus exposures, to speech learning. [NIH supported.]