Can temporal resolution improve with practice? One of the first studies on this issue reported no perceptual learning (Rammsayer 1994). In this study, subjects were trained on 50-ms intervals for 10 min a day for 4 weeks. Subsequent studies revealed robust learning with training (Wright et al. 1997, Nagarajan et al. 1998, Karmarkar & Buonomano 2003). In these studies subjects were generally trained for an hour a day (400–800 trials) for 10 days.

The perceptual learning studies, in addition to suggesting that the neural mechanisms underlying timing can be fine-tuned with experience,

Generalization studies reveal that interval discrimination learning is specific to the temporal domain, and generalization occurs in the spatial domain.

Interval learning has also been reported to generalize across modalities. Nagarajan et al. (1998) show that training on a somatosensory task can produce improvement on an auditory interval discrimination task similar to the interval used for somatosensory training. Even more surprising, training on an auditory task appears to result in an interval-specific improvement in a motor task requiring that the subjects tap their fingers to mark specific intervals (Meegan et al. 2000).

The simplest interpretation of these data is that centralized circuits exist for each interval, and with training, either the temporal accuracy or the downstream processing of these circuits undergoes plasticity. In this interpretation, timing is centralized but interval based. However, it is possible that in these tasks learning occurs as a result of interval-specific cognitive processes other than temporal processing per se. For example, because interval discrimination tasks require comparing the test interval and a standard interval, improvement could rely on better representation of the standard interval or improved storage or retrieval from working or short-term memory. Such alternative explanations would be consistent with the generalization across different stimulus markers and modalities, as well as the lack of generalization to novel intervals. Alternatively, it could be argued that, although many circuits are capable of temporal processing, the relatively simple nature of these temporal tasks allows the brain to use multimodal pathways and a single timing circuit.