Olfactory illusions: Where are they?

Abstract

It has been suggested that there maybe no olfactory illusions. This manuscript examines this claim and argues that it arises because olfactory illusions are not typically accompanied by an awareness of their illusory nature. To demonstrate that olfactory illusions do occur, the relevant empirical literature is reviewed, by examining instances of where the same stimulus results in different percepts, and of where different stimuli result in the same percept. The final part of the manuscript evaluates the evidence favoring the existence of olfactory illusions, and then examines why they may not typically be accompanied by awareness. Three contributory mechanisms are discussed, relating to difficulty of verification and paucity of olfactory knowledge, the role of change blindness, and restricted access consciousness in this sense.

Introduction

A number of authors have suggested that there maybe no olfactory illusions (e.g., Batty, 2010, Lycan, 2000). This assumed absence seems to reflect a broader view held by experimental psychologists. For example, when consulting the index and content of various popular perception textbooks (e.g. Goldstein, 2002, Sekuler and Blake, 2002) or indeed, recent specialist books on olfaction (e.g. Brewer et al., 2006, Doty, 2003, Hawkins and Doty, 2009, Wilson and Stevenson, 2006) – none of these contain a mention of olfactory illusions. This would seem to suggest that there are either no olfactory illusions or they somehow escape notice. Either conclusion would have interesting implications, both specific to current philosophical debates about the status of olfactory experience (e.g. Batty, 2010, Lycan, 2000), and relatedly to the apparent psychological and neural uniqueness of olfaction relative to the other senses (Stevenson, 2009a, Stevenson, 2009b, Smythies, 1997, Verhagen, 2006, Zucco, 2003). More broadly, considerations of human consciousness are highly visuo-centric, but if olfaction is somehow unique, then visually centered approaches may founder when applied to this sense. As this manuscript will show there is good experimental evidence that there are olfactory illusions, but what emerges from this literature is that olfactory illusions seem to lack an interesting feature, which is present for many, but not all illusions in other modalities. This feature concerns awareness of having experienced an illusion.

To evaluate whether there are olfactory illusions and whether we typically lack awareness of them – the manuscript is organized into three parts. The first deals with issues of definition. Unless there is some agreement over what is and what is not meant by the term ‘illusion’, it becomes problematic to determine which senses may demonstrate them. The second part of the manuscript presents an overview of the main classes of illusion that have been documented for olfaction. Items are included here on the basis of definition and on the grounds of similarity with phenomena categorized as illusions in the other sensory systems (e.g. see, Gregory, 1997, Gregory, 2005, Hawkins, 2010, Hayward, 2008). The third part of the manuscript examines why olfactory illusions may be qualitatively different from those experienced in other modalities, especially with regard to awareness. This difference, so it is argued, relates to issues of verification (i.e. ones capacity to independently confirm what one is smelling), variation in olfactory perceptual ability and knowledge, the phenomenon of change blindness, and that access consciousness may be absent or restricted in olfaction. The specific and broader implications of this ‘lack of awareness’ account are then considered.

The word illusion derives from the Latin illusio, meaning “deceit, to mock or make sport with, the saying of the opposite of what is meant” (Glare, 1982). This Latin meaning suggests two concepts, which seem to be relevant to the use of the term illusion – an objective disparity or contradiction, and perhaps less apparently, a potential for subjective awareness of this disparity. This two-fold root to the word illusion is not always apparent in contemporary psychological definitions. Table 1 presents six definitions taken from current dictionaries of psychology that were available in the reference section of the University library (several of the authors here are distinguished experts in perception so these entries are likely to be representative). With the exception of Reber (1985), they define an illusion as a misperception or error – that is a disparity between some objective state of the world and ones perception of it. Reber’s definition is consistent with this interpretation, except that he avoids terming it an error. Perhaps more importantly, with the possible exception of Coleman (2003), and the probable exception of Gregory and Zanwill (1987) – see Table 1 – these definitions do not allude to the issue of a subjective awareness of the disparity or error. On this basis, one could make a consensus definition that an illusion is a misperception, namely a disparity between objective reality and perception. This definition would then correctly classify as illusions, most visual, auditory and somatosensory phenomena that are thought of as illusions by psychologists. However, this definition avoids the issue of subjective awareness of having experienced an illusion. This avoidance may be of no consequence for visual, auditory and tactile illusions. This is because in these senses people may be readily able to detect some illusions themselves (e.g. waterfall illusion [after staring at a waterfall if ones gaze falls upon the rocks at the side, these now appear to be moving ‘up’]), while most others may be detectable once they are pointed out (e.g. change blindness [not noticing the change in some central feature of the stimulus between one point in time and another]). On this basis, one could nuance the claim regarding an absence of awareness of olfactory illusions into three variants, going from: (1) they are typically not noticed but could be noticed if pointed out; to (2) even if pointed out they may be less apparent than for comparable phenomena in other modalities; and to (3) that people may be incapable of noticing them, even if pointed out.

Even if one does not agree with this consensus definition of an illusion as a misperception, it is still possible to adopt a more pragmatic stance to determining whether they exist. Several authors (Gregory, 1997, Gregory, 2005, Hawkins, 2010, Hayward, 2008) have either categorized or listed illusions drawn from the various senses – although no such survey has been attempted before for olfaction. To establish whether there are olfactory illusions one could then simply look for cases that broadly parallel examples drawn from these various sources. This task is made easier by the very useful tables that Gregory, 1997, Gregory, 2005) assembled of various visual illusions, and these have been used here as a pragmatic guide for inclusion. However, this approach also has its limitations. Some olfactory items are included in this manuscript (notably those in the Miscellanea section), which seem relevant here, but that are not illusions either by reference to Gregory’s tables or in terms of the definition adopted above. These items are included because they assist in identifying why olfactory illusions may be qualitatively different. In addition, not all olfactory phenomena, which parallel entries on Gregory’s tables, have been included, as some are rather tangential. Notably, these consist of medically induced perceptual distortions and optical ambiguities, all of which have olfactory parallels, but which do not directly relate to the two claims evaluated here. Table 2 provides a summary list of the olfactory illusions, which are described in the next section, along with their visual analog (where relevant) as identified in Gregory, 1997, Gregory, 2005.