Limits...
Differing effects of attention in single-units and populations are well predicted by heterogeneous tuning and the normalization model of attention.

Hara Y, Pestilli F, Gardner JL - Front Comput Neurosci (2014)

Bottom Line: The normalization model of attention elegantly predicts the diversity of effects of attention reported in single-units well-tuned to the stimulus, but what predictions does it make for more realistic populations of neurons with heterogeneous tuning?We found that within the population, neurons well-tuned to the stimulus showed a response-gain effect, while less-well-tuned neurons showed a contrast-gain effect.More generally, computational models can unify physiological findings across different scales of measurement, and make links to behavior, but only if factors such as heterogeneous tuning within a population are properly accounted for.

View Article: PubMed Central - PubMed

Affiliation: Laboratory for Human Systems Neuroscience, RIKEN Brain Science Institute Wako, Japan.

ABSTRACT
Single-unit measurements have reported many different effects of attention on contrast-response (e.g., contrast-gain, response-gain, additive-offset dependent on visibility), while functional imaging measurements have more uniformly reported increases in response across all contrasts (additive-offset). The normalization model of attention elegantly predicts the diversity of effects of attention reported in single-units well-tuned to the stimulus, but what predictions does it make for more realistic populations of neurons with heterogeneous tuning? Are predictions in accordance with population-scale measurements? We used functional imaging data from humans to determine a realistic ratio of attention-field to stimulus-drive size (a key parameter for the model) and predicted effects of attention in a population of model neurons with heterogeneous tuning. We found that within the population, neurons well-tuned to the stimulus showed a response-gain effect, while less-well-tuned neurons showed a contrast-gain effect. Averaged across the population, these disparate effects of attention gave rise to additive-offsets in contrast-response, similar to reports in human functional imaging as well as population averages of single-units. Differences in predictions for single-units and populations were observed across a wide range of model parameters (ratios of attention-field to stimulus-drive size and the amount of baseline response modifiable by attention), offering an explanation for disparity in physiological reports. Thus, by accounting for heterogeneity in tuning of realistic neuronal populations, the normalization model of attention can not only predict responses of well-tuned neurons, but also the activity of large populations of neurons. More generally, computational models can unify physiological findings across different scales of measurement, and make links to behavior, but only if factors such as heterogeneous tuning within a population are properly accounted for.

No MeSH data available.


Related in: MedlinePlus

Task design. Subjects performed a contrast-discrimination task in one of four locations. On each trial, four contrast gratings appeared in two temporal intervals (Stim1 and Stim2) separated by an inter-stimulus interval (ISI). During one of the two intervals (Stim2 for this figure), the contrast in one location (target, upper-right for this figure) was incremented by a threshold contrast. After both stimulus presentation intervals, a green response-cue indicated the target location and subjects reported the interval during which they perceived the higher contrast with a key press (Resp). At the beginning of each trial (Cue), a white line pointed to one (or, on alternate trials, more than one) of the possible target locations, thus varying the prior information given to subjects regarding which location they would be asked to respond about. Trials were separated by an inter-trial interval (ITI) which lasted 1.5 s for tasks performed outside the scanner and 1.5–12.0 s, pseudo-randomized, inside the scanner.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC3928538&req=5

Figure 1: Task design. Subjects performed a contrast-discrimination task in one of four locations. On each trial, four contrast gratings appeared in two temporal intervals (Stim1 and Stim2) separated by an inter-stimulus interval (ISI). During one of the two intervals (Stim2 for this figure), the contrast in one location (target, upper-right for this figure) was incremented by a threshold contrast. After both stimulus presentation intervals, a green response-cue indicated the target location and subjects reported the interval during which they perceived the higher contrast with a key press (Resp). At the beginning of each trial (Cue), a white line pointed to one (or, on alternate trials, more than one) of the possible target locations, thus varying the prior information given to subjects regarding which location they would be asked to respond about. Trials were separated by an inter-trial interval (ITI) which lasted 1.5 s for tasks performed outside the scanner and 1.5–12.0 s, pseudo-randomized, inside the scanner.

Mentions: Subjects performed a contrast-discrimination task as illustrated in Figure 1. Four contrast gratings (spatial frequency = 2 cycles/°, size = 3° radius, 6° eccentricity in each visual quadrant, contrasts = 12.5, 25, and 50%) were presented in two separate temporal intervals (Stim1 and Stim2, each 600 ms). All stimuli maintained the same contrast in the two intervals except one—the target stimulus, which had a slightly higher contrast in one of the two intervals. After stimulus offset, during the response interval, a green line indicated the location of the target. Subjects were asked to report the interval in which the target had the higher contrast. The difference in contrast presented between the two intervals was adjusted to a threshold level using a 1-up-2-down staircase procedure (Levitt, 1971). The target location and only the target location was shown with a contrast difference between the two temporal intervals. Cues (white line) presented at the beginning of the trial (1 s before stimulus presentation) and throughout the trial (until beginning of response interval, total duration of 2.5 s) indicated which target would need to be discriminated. Subjects were told to attend to cued locations and ignore as much as possible the other locations. In interleaved trials, the cues could point to one, two or four of the locations.


Differing effects of attention in single-units and populations are well predicted by heterogeneous tuning and the normalization model of attention.

Hara Y, Pestilli F, Gardner JL - Front Comput Neurosci (2014)

Task design. Subjects performed a contrast-discrimination task in one of four locations. On each trial, four contrast gratings appeared in two temporal intervals (Stim1 and Stim2) separated by an inter-stimulus interval (ISI). During one of the two intervals (Stim2 for this figure), the contrast in one location (target, upper-right for this figure) was incremented by a threshold contrast. After both stimulus presentation intervals, a green response-cue indicated the target location and subjects reported the interval during which they perceived the higher contrast with a key press (Resp). At the beginning of each trial (Cue), a white line pointed to one (or, on alternate trials, more than one) of the possible target locations, thus varying the prior information given to subjects regarding which location they would be asked to respond about. Trials were separated by an inter-trial interval (ITI) which lasted 1.5 s for tasks performed outside the scanner and 1.5–12.0 s, pseudo-randomized, inside the scanner.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC3928538&req=5

Figure 1: Task design. Subjects performed a contrast-discrimination task in one of four locations. On each trial, four contrast gratings appeared in two temporal intervals (Stim1 and Stim2) separated by an inter-stimulus interval (ISI). During one of the two intervals (Stim2 for this figure), the contrast in one location (target, upper-right for this figure) was incremented by a threshold contrast. After both stimulus presentation intervals, a green response-cue indicated the target location and subjects reported the interval during which they perceived the higher contrast with a key press (Resp). At the beginning of each trial (Cue), a white line pointed to one (or, on alternate trials, more than one) of the possible target locations, thus varying the prior information given to subjects regarding which location they would be asked to respond about. Trials were separated by an inter-trial interval (ITI) which lasted 1.5 s for tasks performed outside the scanner and 1.5–12.0 s, pseudo-randomized, inside the scanner.
Mentions: Subjects performed a contrast-discrimination task as illustrated in Figure 1. Four contrast gratings (spatial frequency = 2 cycles/°, size = 3° radius, 6° eccentricity in each visual quadrant, contrasts = 12.5, 25, and 50%) were presented in two separate temporal intervals (Stim1 and Stim2, each 600 ms). All stimuli maintained the same contrast in the two intervals except one—the target stimulus, which had a slightly higher contrast in one of the two intervals. After stimulus offset, during the response interval, a green line indicated the location of the target. Subjects were asked to report the interval in which the target had the higher contrast. The difference in contrast presented between the two intervals was adjusted to a threshold level using a 1-up-2-down staircase procedure (Levitt, 1971). The target location and only the target location was shown with a contrast difference between the two temporal intervals. Cues (white line) presented at the beginning of the trial (1 s before stimulus presentation) and throughout the trial (until beginning of response interval, total duration of 2.5 s) indicated which target would need to be discriminated. Subjects were told to attend to cued locations and ignore as much as possible the other locations. In interleaved trials, the cues could point to one, two or four of the locations.

Bottom Line: The normalization model of attention elegantly predicts the diversity of effects of attention reported in single-units well-tuned to the stimulus, but what predictions does it make for more realistic populations of neurons with heterogeneous tuning?We found that within the population, neurons well-tuned to the stimulus showed a response-gain effect, while less-well-tuned neurons showed a contrast-gain effect.More generally, computational models can unify physiological findings across different scales of measurement, and make links to behavior, but only if factors such as heterogeneous tuning within a population are properly accounted for.

View Article: PubMed Central - PubMed

Affiliation: Laboratory for Human Systems Neuroscience, RIKEN Brain Science Institute Wako, Japan.

ABSTRACT
Single-unit measurements have reported many different effects of attention on contrast-response (e.g., contrast-gain, response-gain, additive-offset dependent on visibility), while functional imaging measurements have more uniformly reported increases in response across all contrasts (additive-offset). The normalization model of attention elegantly predicts the diversity of effects of attention reported in single-units well-tuned to the stimulus, but what predictions does it make for more realistic populations of neurons with heterogeneous tuning? Are predictions in accordance with population-scale measurements? We used functional imaging data from humans to determine a realistic ratio of attention-field to stimulus-drive size (a key parameter for the model) and predicted effects of attention in a population of model neurons with heterogeneous tuning. We found that within the population, neurons well-tuned to the stimulus showed a response-gain effect, while less-well-tuned neurons showed a contrast-gain effect. Averaged across the population, these disparate effects of attention gave rise to additive-offsets in contrast-response, similar to reports in human functional imaging as well as population averages of single-units. Differences in predictions for single-units and populations were observed across a wide range of model parameters (ratios of attention-field to stimulus-drive size and the amount of baseline response modifiable by attention), offering an explanation for disparity in physiological reports. Thus, by accounting for heterogeneity in tuning of realistic neuronal populations, the normalization model of attention can not only predict responses of well-tuned neurons, but also the activity of large populations of neurons. More generally, computational models can unify physiological findings across different scales of measurement, and make links to behavior, but only if factors such as heterogeneous tuning within a population are properly accounted for.

No MeSH data available.


Related in: MedlinePlus