Cite as:
Rose D, 1998, "'Labelled lines' as a Ryleian category error" Perception 27 ECVP Abstract Supplement
'Labelled lines' as a Ryleian category error
D Rose
The concept of 'labelled lines' is used frequently in modern psychophysics to explain a variety of phenomena, including stimulus identification at detection threshold and perceived stimulus quality. A label may be defined as "a hypothetical 'extra process' that accompanies a sensory message and gives a clue to its origin" [Morgan, 1977 Molyneux's Question (Cambridge: Cambridge University Press) p. 141]. However, the nature of the mysterious 'extra process' has never been clarified.
Sensory information leads to 'representations', which interact with each other as part of the cognitive 'semantic network' and ultimately have a causal role in affecting behaviour. A distinction must, however, be made between the representation (the vehicle or medium) and its content (the meaningful information contained within it) [eg Dretske, 1995 Naturalizing the Mind (Cambridge, MA: MIT Press)]. I propose here that this distinction maps onto that between 'labels' and Morgan's 'sensory messages': the former refers to the medium and the latter to the content. Thus the theory of labelled lines assumes that information is available not only about the content but also about the medium by which the message is conveyed. An example of such use is the idea that information is coded as position in anatomical topographic maps (Konishi, 1990 Cold Spring Harbor Symposia 55 575 - 584).
Given that the medium and the message are two different categories of entity, to treat them both as (equivalently accessible types of 'information' would be to commit a category error, in the sense made famous by Ryle [1949 The Concept of Mind (London: Hutchinson)].
These web-based abstracts are provided for ease of seaching and access, but certain aspects (such as as mathematics) may not appear in their optimum form. For the final published version of this abstract, please see
ECVP 1998 Abstract Supplement (complete) size: 2107 Kb
