Front or Back of the Cortex? The Debate That Decides Which Theory Survives

Abstract

In 2017 the Journal of Neuroscience published two papers side by side, written by opposing groups, on a question that sounds almost geographical: are the neural correlates of consciousness in the front or the back of the cerebral cortex?

It is not a minor dispute. The answer determines which of the major theories of consciousness survives, and both sides marshal lesion data, imaging and decoding evidence that look convincing in isolation.

This article presents both papers at comparable length. Our own view is that the disagreement is partly about evidence and substantially about what counts as evidence — which is why it has not been settled by adding more of it.

1.The Case for the Back

Boly and colleagues argue that posterior cortex — parietal, occipital and temporal regions — specifies the contents of consciousness, and that prefrontal cortex is largely unnecessary for experience itself. Their summary of the opposing position is that evidence for direct, content-specific involvement of the front of the cortex is missing or unclear.

Their most striking evidence is clinical. Patients with extensive bilateral removal of the frontal lobes remained conscious; one such patient was not noticed as abnormal during social interaction, and another showed an increase in IQ after near-complete bilateral prefrontal resection.

Posterior damage behaves entirely differently. Lesions of the posterior corpus callosum were found in 98 per cent of patients remaining in a vegetative state after one year, carrying a 214-fold risk of permanent vegetative state. Damage to specific posterior areas removes specific contents: fusiform lesions abolish face recognition, occipital damage abolishes colour experience, parietal damage abolishes awareness of one's own movement.

Imaging points the same way. Dream studies identified a posterior hot zone in parieto-occipital areas as a reliable correlate, and specific dream contents — faces, places, movement, speech — could be predicted from posterior but not anterior cortex.

The most elegant piece is a reverse-inference analysis. Asking not what activates during consciousness experiments but how predictive prefrontal activity is of consciousness research specifically, frontal cortex disappears: the prefrontal coordinate turned out to be predictive of the terms 'phonological' and 'task' instead.

2.The Report Problem

Underlying the posterior case is a methodological argument that has changed how these experiments are designed.

Classic paradigms ask participants to report what they perceived, which means every measurement confounds being conscious of something with reporting it. In no-report paradigms, consciously experienced but task-irrelevant stimuli activated posterior regions, while frontal activation correlated with reporting rather than with experience.

If that holds, decades of prefrontal findings may index the machinery of pressing a button and describing an experience rather than the experience itself. It is a serious charge and it is why the no-report approach has spread.

3.The Case for the Front

Odegaard, Knight and Lau reply that the posterior case rests on selective citation of studies using methods with limited sensitivity, and that individual null findings obtained with one measure may not generalise to other measures.

Their statistical point is the strongest part. Prefrontal cortex uses distributed coding, so univariate analyses may miss what multivariate decoding detects, and lesion effects may look modest because many neurons across large prefrontal regions would need to be affected before a deficit appears. Absence of evidence, again, being distinguished from evidence of absence.

Their clinical evidence points the other way from Boly's. Bilateral lateral prefrontal lesions produce severe abulia and disconnection from the environment, unlike occipital or parietal lesions which preserve general awareness. Prefrontal lesions reduced subjective confidence ratings by nearly half while leaving objective task performance intact — a dissociation hard to explain if prefrontal cortex only handles reporting.

They add decoding results: multivariate imaging shows prefrontal activity reflecting specific perceptual content, and single-unit recordings in monkeys found prefrontal activity reflecting an unreported feature of a stimulus even when no report was required.

On no-report paradigms they are dismissive, arguing the excitement may be misguided since prefrontal involvement had already survived such tests before their recent advocacy.

4.What Each Side Concedes

Both papers contain concessions that are more informative than their headline claims.

Boly and colleagues allow that some prefrontal regions — ventromedial or premotor areas — may contribute specific conscious contents such as feelings of reflection, valuation and affect. So the posterior claim is about perceptual content rather than about everything.

Odegaard and colleagues concede that incomplete unilateral lesions often produce subtle or null effects, that prefrontal decoding accuracy is limited in general, and that some recent no-report studies do find reduced prefrontal involvement. They explicitly do not claim that prefrontal cortex fully determines perceptual content, proposing instead that it works together with early visual areas where the specific content resides.

Read the concessions together and the gap narrows considerably. Posterior regions carry perceptual content; prefrontal regions contribute something — confidence, valuation, access — whose relationship to experience is what remains genuinely unresolved.

5.What Happened Next

The adversarial collaboration described elsewhere in this library was designed in part to break this deadlock, and its findings bear on it directly. Adding prefrontal regions did not improve decoding of stimulus category or orientation, and in some analyses reduced it. Cross-temporal generalisation was sustained in posterior regions and brief in prefrontal ones.

That is a point for the posterior side on this specific measure, obtained under conditions both camps agreed to in advance. It did not end the argument — the same study raised problems for the posterior theory's connectivity predictions — but it is the closest thing to a neutral referee this dispute has had.

6.How We Read It

Our reading is that Boly and colleagues have the stronger clinical evidence and Odegaard and colleagues the stronger methodological caution, and that both are right about the other's weakness.

The deeper problem is that the two camps disagree about what would count as settling it. If prefrontal activity during reporting is a confound, prefrontal findings are artefacts. If no-report paradigms simply lack sensitivity, posterior findings are incomplete. Each side's methodological principle makes the other's best evidence inadmissible, and no quantity of additional data resolves a disagreement of that shape.

Which is precisely why the field turned to adversarial collaboration — agreeing on admissibility before collecting anything is the only move available when the argument is about the rules rather than the results.

Editorial Comment

MindHeaven® makes no claim relating to consciousness and nothing here concerns any product. We have given both papers roughly equal space deliberately; a summary that presented only the posterior case would have been shorter, cleaner and misleading.