Limits...
Sensory adaptation for timing perception.

Roseboom W, Linares D, Nishida S - Proc. Biol. Sci. (2015)

Bottom Line: Here, we show that the effect of recent experience on timing perception is not just subjective; recent sensory experience also modifies relative timing discrimination.This result indicates that recent sensory history alters the encoding of relative timing in sensory areas, excluding explanations of the subjective phenomenon based only on decision-level changes.The existence of these components would suggest that previous explanations of how recent experience may change the sensory encoding of timing, such as changes in sensory latencies or simple implementations of neural population codes, cannot account for the effect of sensory adaptation on timing perception.

View Article: PubMed Central - PubMed

Affiliation: NTT Communication Science Laboratories, Nippon Telegraph and Telephone Corporation, 3-1 Morino-sato Wakamiya, Atsugi-shi, Kanagawa, 243-0198, Japan wjroseboom@gmail.com.

ABSTRACT
Recent sensory experience modifies subjective timing perception. For example, when visual events repeatedly lead auditory events, such as when the sound and video tracks of a movie are out of sync, subsequent vision-leads-audio presentations are reported as more simultaneous. This phenomenon could provide insights into the fundamental problem of how timing is represented in the brain, but the underlying mechanisms are poorly understood. Here, we show that the effect of recent experience on timing perception is not just subjective; recent sensory experience also modifies relative timing discrimination. This result indicates that recent sensory history alters the encoding of relative timing in sensory areas, excluding explanations of the subjective phenomenon based only on decision-level changes. The pattern of changes in timing discrimination suggests the existence of two sensory components, similar to those previously reported for visual spatial attributes: a lateral shift in the nonlinear transducer that maps relative timing into perceptual relative timing and an increase in transducer slope around the exposed timing. The existence of these components would suggest that previous explanations of how recent experience may change the sensory encoding of timing, such as changes in sensory latencies or simple implementations of neural population codes, cannot account for the effect of sensory adaptation on timing perception.

Show MeSH

Related in: MedlinePlus

Lateral shift plus repulsion. (a) Depiction of the model. (b) Sensitivity as a function of the asynchrony for participants YI (left) and WR (right) for the different conditions. These data are the same as plotted in figure 2. The black curves correspond to transducers in equation (2.1) that best fitted the data for the no adaptation condition. The coloured curves correspond to the curves for lateral shift-plus-repulsion that best fitted the data for the different adaptation conditions. (c) Best parameter for the lateral shift-plus-repulsion model for each participant and for the average across participants for the different adaptation conditions. The error bars correspond to the within-subjects 95% CIs calculated according to Morey [18].
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4389610&req=5

RSPB20142833F5: Lateral shift plus repulsion. (a) Depiction of the model. (b) Sensitivity as a function of the asynchrony for participants YI (left) and WR (right) for the different conditions. These data are the same as plotted in figure 2. The black curves correspond to transducers in equation (2.1) that best fitted the data for the no adaptation condition. The coloured curves correspond to the curves for lateral shift-plus-repulsion that best fitted the data for the different adaptation conditions. (c) Best parameter for the lateral shift-plus-repulsion model for each participant and for the average across participants for the different adaptation conditions. The error bars correspond to the within-subjects 95% CIs calculated according to Morey [18].

Mentions: We assessed whether the combination of the two previous models (figure 5a,b) provided a better fit of the changes in sensitivity caused by adaptation (electronic supplementary material, figure S1). The lateral-shift-plus-repulsion model in comparison with the lateral shift model improved the fit for all conditions: VA adaptation (t7 = 2.47, p = 0.042); AV adaptation (t7 = 3.63, p = 0.0084) and synchrony adaptation (t7 = 3.042, p = 0.019). The lateral-shift-plus-repulsion model in comparison with the repulsion model improved the fit for the asynchrony adaptation conditions: VA adaptation (t7 = 4.05, p = 0.0049) and AV adaptation (t7 = 3.50, p = 0.010). It did not improve the fit for the synchrony adaptation condition (t7 = 1.71, p = 0.13). The improvements were not just owing to an increase in the number of the free parameters as, when combining all conditions, the Akaike information criterion (see the electronic supplementary material, Methods) for all participants was lower for the lateral shift-plus-repulsion model indicating that it is a more probable model (see the electronic supplementary material, figure S1). Confirming the superiority of the lateral shift-plus-repulsion model, likelihood ratio tests (see the electronic supplementary material, Methods) for the lateral shift and repulsion models relative to the lateral shift-plus-repulsion were highly significant for all participants (the largest probability was for participant BO for the comparison of repulsion versus lateral shift-plus-repulsion: p = 0.02, χ1 = 5.41). As for the two parameters of the combined model, they were very similar to the parameters obtained by independent fits of lateral shift and repulsion (figure 5c,d).Figure 5.


Sensory adaptation for timing perception.

Roseboom W, Linares D, Nishida S - Proc. Biol. Sci. (2015)

Lateral shift plus repulsion. (a) Depiction of the model. (b) Sensitivity as a function of the asynchrony for participants YI (left) and WR (right) for the different conditions. These data are the same as plotted in figure 2. The black curves correspond to transducers in equation (2.1) that best fitted the data for the no adaptation condition. The coloured curves correspond to the curves for lateral shift-plus-repulsion that best fitted the data for the different adaptation conditions. (c) Best parameter for the lateral shift-plus-repulsion model for each participant and for the average across participants for the different adaptation conditions. The error bars correspond to the within-subjects 95% CIs calculated according to Morey [18].
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4389610&req=5

RSPB20142833F5: Lateral shift plus repulsion. (a) Depiction of the model. (b) Sensitivity as a function of the asynchrony for participants YI (left) and WR (right) for the different conditions. These data are the same as plotted in figure 2. The black curves correspond to transducers in equation (2.1) that best fitted the data for the no adaptation condition. The coloured curves correspond to the curves for lateral shift-plus-repulsion that best fitted the data for the different adaptation conditions. (c) Best parameter for the lateral shift-plus-repulsion model for each participant and for the average across participants for the different adaptation conditions. The error bars correspond to the within-subjects 95% CIs calculated according to Morey [18].
Mentions: We assessed whether the combination of the two previous models (figure 5a,b) provided a better fit of the changes in sensitivity caused by adaptation (electronic supplementary material, figure S1). The lateral-shift-plus-repulsion model in comparison with the lateral shift model improved the fit for all conditions: VA adaptation (t7 = 2.47, p = 0.042); AV adaptation (t7 = 3.63, p = 0.0084) and synchrony adaptation (t7 = 3.042, p = 0.019). The lateral-shift-plus-repulsion model in comparison with the repulsion model improved the fit for the asynchrony adaptation conditions: VA adaptation (t7 = 4.05, p = 0.0049) and AV adaptation (t7 = 3.50, p = 0.010). It did not improve the fit for the synchrony adaptation condition (t7 = 1.71, p = 0.13). The improvements were not just owing to an increase in the number of the free parameters as, when combining all conditions, the Akaike information criterion (see the electronic supplementary material, Methods) for all participants was lower for the lateral shift-plus-repulsion model indicating that it is a more probable model (see the electronic supplementary material, figure S1). Confirming the superiority of the lateral shift-plus-repulsion model, likelihood ratio tests (see the electronic supplementary material, Methods) for the lateral shift and repulsion models relative to the lateral shift-plus-repulsion were highly significant for all participants (the largest probability was for participant BO for the comparison of repulsion versus lateral shift-plus-repulsion: p = 0.02, χ1 = 5.41). As for the two parameters of the combined model, they were very similar to the parameters obtained by independent fits of lateral shift and repulsion (figure 5c,d).Figure 5.

Bottom Line: Here, we show that the effect of recent experience on timing perception is not just subjective; recent sensory experience also modifies relative timing discrimination.This result indicates that recent sensory history alters the encoding of relative timing in sensory areas, excluding explanations of the subjective phenomenon based only on decision-level changes.The existence of these components would suggest that previous explanations of how recent experience may change the sensory encoding of timing, such as changes in sensory latencies or simple implementations of neural population codes, cannot account for the effect of sensory adaptation on timing perception.

View Article: PubMed Central - PubMed

Affiliation: NTT Communication Science Laboratories, Nippon Telegraph and Telephone Corporation, 3-1 Morino-sato Wakamiya, Atsugi-shi, Kanagawa, 243-0198, Japan wjroseboom@gmail.com.

ABSTRACT
Recent sensory experience modifies subjective timing perception. For example, when visual events repeatedly lead auditory events, such as when the sound and video tracks of a movie are out of sync, subsequent vision-leads-audio presentations are reported as more simultaneous. This phenomenon could provide insights into the fundamental problem of how timing is represented in the brain, but the underlying mechanisms are poorly understood. Here, we show that the effect of recent experience on timing perception is not just subjective; recent sensory experience also modifies relative timing discrimination. This result indicates that recent sensory history alters the encoding of relative timing in sensory areas, excluding explanations of the subjective phenomenon based only on decision-level changes. The pattern of changes in timing discrimination suggests the existence of two sensory components, similar to those previously reported for visual spatial attributes: a lateral shift in the nonlinear transducer that maps relative timing into perceptual relative timing and an increase in transducer slope around the exposed timing. The existence of these components would suggest that previous explanations of how recent experience may change the sensory encoding of timing, such as changes in sensory latencies or simple implementations of neural population codes, cannot account for the effect of sensory adaptation on timing perception.

Show MeSH
Related in: MedlinePlus