Limits...
Preferential encoding of behaviorally relevant predictions revealed by EEG.

Stokes MG, Myers NE, Turnbull J, Nobre AC - Front Hum Neurosci (2014)

Bottom Line: In this electroencephalogram (EEG) study, we test how task relevance influences the way predictions are learned from the statistics of visual input, and exploited for behavior.The behavioral results confirmed that participants learned and exploited task-relevant predictions even when not explicitly defined.These results show that task relevance critically influences how the brain extracts predictive structure from the environment, and exploits these regularities for optimized behavior.

View Article: PubMed Central - PubMed

Affiliation: Department of Experimental Psychology, University of Oxford Oxford, UK ; Oxford Centre for Human Brain Activity, University of Oxford Oxford, UK.

ABSTRACT
Statistical regularities in the environment guide perceptual processing; however, some predictions are bound to be more important than others. In this electroencephalogram (EEG) study, we test how task relevance influences the way predictions are learned from the statistics of visual input, and exploited for behavior. We developed a novel task in which participants are simply instructed to respond to a designated target stimulus embedded in a serial stream of non-target stimuli. Presentation probabilities were manipulated such that a designated target cue stimulus predicted the target onset with 70% validity. We also included a corresponding control contingency: a pre-designated control cue predicted a specific non-target stimulus with 70% validity. Participants were not informed about these contingencies. This design allowed us to examine the neural response to task-relevant predictive (cue) and predicted stimuli (target), relative to task-irrelevant predictive (control cue) and predicted stimuli (control non-target). The behavioral results confirmed that participants learned and exploited task-relevant predictions even when not explicitly defined. The EEG results further showed that target-relevant predictions are coded more strongly than statistically equivalent regularities between non-target stimuli. There was a robust modulation of the response for predicted targets associated with learning, enhancing the response to cued stimuli just after 200 ms post-stimulus in central and posterior electrodes, but no corresponding effects for predicted non-target stimuli. These effects of target prediction were preceded by a sustained frontal negativity following presentation of the predictive cue stimulus. These results show that task relevance critically influences how the brain extracts predictive structure from the environment, and exploits these regularities for optimized behavior.

No MeSH data available.


Related in: MedlinePlus

Event-related potentials to predictive stimuli: target cue and control non-target cues. (A) Plots show the mean potential difference between target cues and neutral stimuli, separately for each block, for the frontal, central, and posterior ROIs (B) Topography of learning effect. The topography shows the mean slope derived from the linear regression of task block onto potential difference. (C) and (D) show the same as (A) and (B), but for the control cue. (E) The mean regression slope across the eight task blocks (fit separately at each time point) is shown for the frontal ROIs shown in (A) and (C), for target cues (blue lines) and control cues (red lines). Shading indicates SEM. There was a significant effect of target cue learning (588–780 ms, p = 0.0343, cluster-corrected).
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4151094&req=5

Figure 4: Event-related potentials to predictive stimuli: target cue and control non-target cues. (A) Plots show the mean potential difference between target cues and neutral stimuli, separately for each block, for the frontal, central, and posterior ROIs (B) Topography of learning effect. The topography shows the mean slope derived from the linear regression of task block onto potential difference. (C) and (D) show the same as (A) and (B), but for the control cue. (E) The mean regression slope across the eight task blocks (fit separately at each time point) is shown for the frontal ROIs shown in (A) and (C), for target cues (blue lines) and control cues (red lines). Shading indicates SEM. There was a significant effect of target cue learning (588–780 ms, p = 0.0343, cluster-corrected).

Mentions: Next, we compared ERPs to the predictive stimuli: target-related cues and cues for control non-target stimuli. Again, plotting the ERPs over blocks throughout the experimental session, there is evidence for the development of a sustained negativity in frontal sensors by the end of the session (Figures 4A,B). This effect was evident as a statistically significant cluster in the regression of cue effect against block number that emerges just before 600 ms in frontal sensors (Figure 4E, 588–780 ms, p = 0.0514). Note, the negative relationship here is consistent with a negative cueing effect (cue < neutral) that increases in magnitude (i.e. gets more negative) over the course of the session (as illustrated in Figure 4A). The regression analysis did not reveal any other significant effects of the task-relevant target cue in the other electrode clusters. No significant effects were observed for the control cue (Figures 4C,D). The difference between the regression slope for target cues and control cues showed a trend, but was not significant (mean over 588–780 ms, t17 = -1.84, p = 0.083). Post hoc follow-up analysis of this frontal cueing effect revealed a significant correlation to behavior (r(17) = 0.582, p = 0.011).


Preferential encoding of behaviorally relevant predictions revealed by EEG.

Stokes MG, Myers NE, Turnbull J, Nobre AC - Front Hum Neurosci (2014)

Event-related potentials to predictive stimuli: target cue and control non-target cues. (A) Plots show the mean potential difference between target cues and neutral stimuli, separately for each block, for the frontal, central, and posterior ROIs (B) Topography of learning effect. The topography shows the mean slope derived from the linear regression of task block onto potential difference. (C) and (D) show the same as (A) and (B), but for the control cue. (E) The mean regression slope across the eight task blocks (fit separately at each time point) is shown for the frontal ROIs shown in (A) and (C), for target cues (blue lines) and control cues (red lines). Shading indicates SEM. There was a significant effect of target cue learning (588–780 ms, p = 0.0343, cluster-corrected).
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4151094&req=5

Figure 4: Event-related potentials to predictive stimuli: target cue and control non-target cues. (A) Plots show the mean potential difference between target cues and neutral stimuli, separately for each block, for the frontal, central, and posterior ROIs (B) Topography of learning effect. The topography shows the mean slope derived from the linear regression of task block onto potential difference. (C) and (D) show the same as (A) and (B), but for the control cue. (E) The mean regression slope across the eight task blocks (fit separately at each time point) is shown for the frontal ROIs shown in (A) and (C), for target cues (blue lines) and control cues (red lines). Shading indicates SEM. There was a significant effect of target cue learning (588–780 ms, p = 0.0343, cluster-corrected).
Mentions: Next, we compared ERPs to the predictive stimuli: target-related cues and cues for control non-target stimuli. Again, plotting the ERPs over blocks throughout the experimental session, there is evidence for the development of a sustained negativity in frontal sensors by the end of the session (Figures 4A,B). This effect was evident as a statistically significant cluster in the regression of cue effect against block number that emerges just before 600 ms in frontal sensors (Figure 4E, 588–780 ms, p = 0.0514). Note, the negative relationship here is consistent with a negative cueing effect (cue < neutral) that increases in magnitude (i.e. gets more negative) over the course of the session (as illustrated in Figure 4A). The regression analysis did not reveal any other significant effects of the task-relevant target cue in the other electrode clusters. No significant effects were observed for the control cue (Figures 4C,D). The difference between the regression slope for target cues and control cues showed a trend, but was not significant (mean over 588–780 ms, t17 = -1.84, p = 0.083). Post hoc follow-up analysis of this frontal cueing effect revealed a significant correlation to behavior (r(17) = 0.582, p = 0.011).

Bottom Line: In this electroencephalogram (EEG) study, we test how task relevance influences the way predictions are learned from the statistics of visual input, and exploited for behavior.The behavioral results confirmed that participants learned and exploited task-relevant predictions even when not explicitly defined.These results show that task relevance critically influences how the brain extracts predictive structure from the environment, and exploits these regularities for optimized behavior.

View Article: PubMed Central - PubMed

Affiliation: Department of Experimental Psychology, University of Oxford Oxford, UK ; Oxford Centre for Human Brain Activity, University of Oxford Oxford, UK.

ABSTRACT
Statistical regularities in the environment guide perceptual processing; however, some predictions are bound to be more important than others. In this electroencephalogram (EEG) study, we test how task relevance influences the way predictions are learned from the statistics of visual input, and exploited for behavior. We developed a novel task in which participants are simply instructed to respond to a designated target stimulus embedded in a serial stream of non-target stimuli. Presentation probabilities were manipulated such that a designated target cue stimulus predicted the target onset with 70% validity. We also included a corresponding control contingency: a pre-designated control cue predicted a specific non-target stimulus with 70% validity. Participants were not informed about these contingencies. This design allowed us to examine the neural response to task-relevant predictive (cue) and predicted stimuli (target), relative to task-irrelevant predictive (control cue) and predicted stimuli (control non-target). The behavioral results confirmed that participants learned and exploited task-relevant predictions even when not explicitly defined. The EEG results further showed that target-relevant predictions are coded more strongly than statistically equivalent regularities between non-target stimuli. There was a robust modulation of the response for predicted targets associated with learning, enhancing the response to cued stimuli just after 200 ms post-stimulus in central and posterior electrodes, but no corresponding effects for predicted non-target stimuli. These effects of target prediction were preceded by a sustained frontal negativity following presentation of the predictive cue stimulus. These results show that task relevance critically influences how the brain extracts predictive structure from the environment, and exploits these regularities for optimized behavior.

No MeSH data available.


Related in: MedlinePlus