Interpretive Bias in Insomnia Introduction Chronic Insomnia Is a Highly Prevalent Disor- Der That Is Associated with Significant Impair- Ment and Disability

CHRONIC INSOMNIA IS A HIGHLY PREVALENT DISORDER THAT IS ASSOCIATED WITH SIGNIFICANT IMPAIRMENT AND DISABILITY (E.G., SEE REFERENCE 1). There has been growing interest in the role of cognitive processes that maintain insomnia (e.g., see references 2-6). However, to date, a role for interpretative bias has not been investigated What is an interpretative bias? Many everyday events are ambiguous and can be interpreted in more than 1 way. Individuals with psychological disorders display a disproportionate tendency to make biased interpretations of ambiguous material, and this bias has been implicated in their maintenance.7,8 An interpretive bias in insomnia would be evident if individuals with insomnia interpreted ambiguous information in an insomnia-consistent manner. For example, the sentence “Melinda thought with anticipation about going to sleep that night” could be interpreted as Melinda worrying about her sleep (an insomnia-consistent interpretation) or awaiting her bedtime eagerly (not an insomniaconsistent interpretation). The present study aims to begin the process of determining whether an interpretive bias exists in poor sleepers. Because this is the first exploration of interpretative bias in the context of insomnia, it was deemed appropriate to use an analog sample. An analog sample in this context refers to a sample that does not include patients, but which includes an approximation of them; in this case., a group of poor sleepers. The rationale for taking this approach is (1) it is plausible that the process of interest exists on a continuum whereby patients who meet full diagnostic criteria for insomnia and poor sleepers both exhibit the process but that the process is present in a stronger form among insomnia patients (e.g., see Flett et al9 for evidence pertaining to this assumption in the depression literature) and (2) the ease of recruiting an analog sample, relative to a patient sample, render more complex experimental designs with larger samples achievable.10 We suggest that this is a reasonable first step on the basis that every science needs an analog to test and hone new hypotheses in a timely and resource-efficient manner.10 Having said that, if the results produced from our poor-sleeper sample are promising, it will be essential to recruit a sample that meets full diagnostic criteria for insomnia to check the generalizability of the findings. The prediction tested was that the poor-sleeper group, relative to the normal-sleeper group, would show an enhanced tendency to interpret ambiguous material in an insomnia-consistent manner. Given the potential for normalor poor-sleeper status to be confounded with anxiety, trait anxiety was included as an independent variable. This allowed for the contribution of insomnia and anxiety on interpretation bias to be independently assessed. A subsidiary aim of the current study was to assess the specificity of interpretive bias in insomnia. To achieve this aim we included ambiguous anxiety-related material in addition to ambiguous insomnia-related material. This permitted us to examine whether any interpretive bias observed is specific to insomnia or also present for anxiety-related information. An Investigation of Interpretive Bias in Insomnia: An Analog Study Comparing Normal and Poor Sleepers