Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 19 (5), 1175-85

A Dual Role for Prediction Error in Associative Learning

Affiliations

A Dual Role for Prediction Error in Associative Learning

Hanneke E M den Ouden et al. Cereb Cortex.

Abstract

Confronted with a rich sensory environment, the brain must learn statistical regularities across sensory domains to construct causal models of the world. Here, we used functional magnetic resonance imaging and dynamic causal modeling (DCM) to furnish neurophysiological evidence that statistical associations are learnt, even when task-irrelevant. Subjects performed an audio-visual target-detection task while being exposed to distractor stimuli. Unknown to them, auditory distractors predicted the presence or absence of subsequent visual distractors. We modeled incidental learning of these associations using a Rescorla-Wagner (RW) model. Activity in primary visual cortex and putamen reflected learning-dependent surprise: these areas responded progressively more to unpredicted, and progressively less to predicted visual stimuli. Critically, this prediction-error response was observed even when the absence of a visual stimulus was surprising. We investigated the underlying mechanism by embedding the RW model into a DCM to show that auditory to visual connectivity changed significantly over time as a function of prediction error. Thus, consistent with predictive coding models of perception, associative learning is mediated by prediction-error dependent changes in connectivity. These results posit a dual role for prediction-error in encoding surprise and driving associative plasticity.

Figures

Figure 1.
Figure 1.
Experimental design. (A) stimuli presented during the experiment. The “distractor” stimuli, whose associations are being learned incidentally, comprised 2 auditory CS corresponding to high- and low-frequency tones and one visual US consisting of 3 concentric squares. The target stimuli, to which the subjects responded, comprised a white noise burst and a circle. (B) Temporal sequence of a single trial. The CS and US could be either presented or omitted. The average trial duration was 2 s. The TO cue was a small central dot (100 ms); the auditory CS was presented for 500 ms, starting 400 ms after TO. The visual stimulus was presented 750 ms after TO, also for 500 ms. The intertrial interval (ITI) was jittered, ranging from 350–1350 ms, and target stimuli were inserted only in the longest ITIs, lasting for 300 ms.
Figure 2.
Figure 2.
Probabilistic relationship between auditory and visual stimuli. Contingency tables showing the proportion of each trial type occurring during CS+ and CS blocks respectively. Below the tables are the resulting conditional probabilities of the visual stimulus being present (or absent), given the presence (or absence) of the auditory CS; these probabilities can be inferred by comparing the frequencies within each column of the table.
Figure 3.
Figure 3.
Compound learning curves. Learning curves were calculated separately for trials on which the auditory CS was present (dots) and absent (crosses), during CS+ (blue), and CS (red) blocks. Note that learning is slower in the absence of an auditory CS than in its presence and faster for CS+ than for CS trials.
Figure 4.
Figure 4.
Dynamic causal models of learning effects on audio-visual connectivity. For all 3 models, the primary auditory (A1) and visual (V1) areas are both driven by their respective sensory inputs. The first model tested had a single connection from A1 to V1 (M1). In model 2 (M2) the V1 → A1 connection was added. In both M1 and M2, the A1 → V1 connection was allowed to change during CS+ trials as a function of the visual outcome (V+ vs. V) and the RW learning curve (ϕ). This modulatory effect corresponds to the interaction of the auditory CS+ prediction with the visual outcome and models a learning-dependent contribution to V1 responses from CS+ responses in A1; and this contribution depends on whether the visual stimulus was present or not (c.f., a prediction error mediated by top-down signals from A1). In the third model, suggested as a control by one of the reviewers, instead of the A1 → V1 connection, the V1 → A1 connection is modulated by the learning signal.
Figure 5.
Figure 5.
fMRI results. (A) Significant activations in V1 as a function of RW learning, for both the 4-way interaction (CS type × CS presence × visual outcome × RW learning; red), and the simple (3-way) interaction (blue), which is restricted to the CS+ trials (x = −6, also showing the caudate activation) and (B) in the putamen bilaterally (y = 6), displayed on the mean structural image across all subjects. (C) z = 12. Significant 3-way interaction CS type × CS presence × RW learning in the DLPFC and left putamen (red). This interaction is driven by the CS+ trials, as shown by the simple interaction in blue.
Figure 6.
Figure 6.
Learning effects on audio-visual connectivity. Bayesian model comparison showed that the DCM with a single connection from A1 to V1 was superior to the other models. Across subjects, there was a significant “endogenous” or “fixed” strength of the A1 → V1 connection (0.10 s−1, P = 0.003) and a significant learning-induced modulation (magenta arrows) of this connection (P = 0.028). The insets show the parameter estimates for the main effects in both A1 and peripheral V1. The magenta arrows indicate how the main effect in peripheral V1 is modulated by changes in connectivity from A1 to V1 during CS+ trials: over time the response to surprising visual outcomes is upregulated, whereas the response to unsurprising visual outcomes is downregulated. Note that in this plot the magenta arrows designate the direction in which V1 responses change due to modulation of connectivity; for quantitative information on this modulatory effect, see the main text.

Similar articles

See all similar articles

Cited by 110 articles

See all "Cited by" articles

References

    1. Anderson JL, Hutton C, Ashburner J, Turner R, Friston K. Modeling geometric deformations in EPI time series. Neuroimage. 2001;13:903–919. - PubMed
    1. Aron AR, Shohamy D, Clark J, Myers C, Gluck MA, Poldrack RA. Human midbrain sensitivity to cognitive feedback and uncertainty during classification learning. J Neurophysiol. 2004;92:1144–1152. - PubMed
    1. Baier B, Kleinschmidt A, Muller NG. Cross-modal processing in early visual and auditory cortices depends on expected statistical relationship of multisensory information. J Neurosci. 2006;26:12260–12265. - PMC - PubMed
    1. Baldeweg T. Repetition effects to sounds: evidence for predictive coding in the auditory system. Trends Cogn Sci. 2006;10:93–94. - PubMed
    1. Blakemore SJ, Wolpert DM, Frith CD. Central cancellation of self-produced tickle sensation. Nat Neurosci. 1998;1:635–640. - PubMed

Publication types

Feedback