In this appendix, we explore the issue of the brain's coding system in more depth. As noted in the main text, Uttal (2016) makes a key distinction between signs and codes. Signs are merely neurophysiological phenomena that appear to correlate with psychological processes; a code is the actual mechanism by means by which the brain ‘works’. The event-related potential (ERP), the subject of the first author's PhD research, provides a pertinent example of this distinction. The ERP is a slow perturbation in the ongoing EEG activity of the brain which is evoked by a single stimulus, such as a click or a flash. A typical ERP is shown in figure 5: it consists of a series of waves which unfold over the half second or so (500 ms) after the stimulus. By convention, negative is shown upwards, and the waves are accordingly known as P1, N1, and so on.

At that time, the dominant model of human cognition was to see the brain as an information processing system, typically represented as a block diagram involving a number of modules: for perception, attention and memory (figure 5). It was naturally tempting to think of the block diagram as a functional model of mind–brain architecture, and when juxtaposed with the ERP to speculate that perhaps the orderly progression of peaks and troughs in the waveform reflects the sequential activation of ‘cognitive modules’ in the brain.

The PhD examined the relationship between the amplitude of the N1–P2 complex of the ERP and the amount of information conveyed by the stimulus. In a key experiment, the research showed that when the quantity of (temporal) information in two stimuli was equated, no amplitude effects were found (Wastell, 1979a). The evidence was seemingly convincing. However, in that experiment, there was a further condition; the subjects were asked to relax and ignore the stimuli completely, rather than produce a reaction time response as in the main condition. Under these circumstances, the amplitude of N1–P2 was actually greater, even though no information was being processed at all. The seductively appealing idea that the amplitude of peaks in ERP waveform reflects the degree of information processing by ‘brain structures’ was thus contradicted by an inconvenient empirical result.