The patients Cornish describes probably have the post-traumatic stress disorder-like secondary agoraphobia (also known as conditioned, acquired, post-traumatic, memory–trace–overconsolidation-based agoraphobia), not the rarer primary (also known as innate prepotentiated) agoraphobia.

As we have proposed (Reference BrachaBracha, 2006; Reference Bracha, Lenze and SheltonBracha et al, 2006a ), both primary and secondary agoraphobia should be taken out of the panic disorder section of DSM–V/ICD–11. Primary agoraphobia (i.e. with no prior Criterion A traumatic event) should be categorised with the other innate specific phobias and secondary (post-traumatic) agoraphobia should be categorised alongside (a more narrowly defined) post-traumatic stress disorder (PTSD) in a new sub-category of anxiety/stress/fear disorders entitled ‘overconsolidational fear disorders.’ Contrary to dogma, evolutionary hypotheses are testable (Reference Bracha, Yoshioka and MasukawaBracha et al, 2005; Reference BrachaBracha, 2006; Reference Bracha and HayashiBracha & Hayashi, 2006; Reference Bracha, Bienvenu and EatonBracha et al, 2006b ).

Panic attacks away from home are not the sole Criterion A event which (if untreated) are often followed by secondary (memory–trace–overconsolidation-based) agoraphobia. Secondary agoraphobia frequently follows embarrassing experiences away from home, related to psychiatric and non-psychiatric conditions such as chronic motor or vocal tic disorder, trichotillomania, narcolepsy, grand mal seizures, etc. Criterion A experiences (‘events’) such as being bullied, ridiculed, threatened or physically assaulted by school or neighbourhood peers are also often followed by the PTSD-like secondary agoraphobia.

Another diagnosis Cornish should consider is social phobia (i.e. innate fear of simultaneous visual scrutiny by a large group of strangers). During much of the human era of evolutionary adaptedness, being stared at by a large group of non-smiling, non-kin conspecifics was more likely than not to be followed by negative consequences (Reference BrachaBracha, 2006). Evolution is not forward looking and could not anticipate a future where being stared at by a large group of non-smiling strangers might be followed by receiving an honorarium rather than by injury or death.

Most importantly, both the dimensional and categorical approaches planned for DSM–V/ICD–11 should include an evolution-inspired ‘innateness axis’ modelled on Axis V (the global assessment of functioning axis). The dimensional innateness axis score would reflect the clinician's estimate (based on past psychiatric history, genetic history, age at onset, etc.) of the likely ‘hardwiredness’ of a particular patient's symptoms (with low scores indicating a mostly post-traumatic, overconsolidational aetiology and high scores indicating a mostly evolutionarily hardwired aetiology).