Qualitative changes in the experience of self in relation to others (i.e. the social self) have been consistently reported to be a feature of the “silent” initial part of the prodrome Reference Moller and Husby[4] and vulnerability to schizophrenia Reference Klosterkotter, Ebel, Schultze-Lutter and Steinmeyer[2,Reference Stanghellini8]. For example, increased interpersonal sensitivity, or changes in the felt naturalness of social interaction, often precedes behavioural manifestations such as social withdrawal and attenuated negative symptoms Reference Lencz, Smith, Auther, Correll and Cornblatt[3]. Thus, early experiential disturbances of social interaction might be an important early indicator of vulnerability to psychosis in help-seeking subjects.

The recent report by Salokangas et al. Reference Salokangas, Patterson, Heinimaa, Svirskis, From and Vaskelainen[6] offers crucial empirical evidence that seems to support this idea. Specifically, it shows that a not-yet psychotic, paranoia-tinged social sensitivity, consisting of the subjective experience of negative attitude of others (NAO) towards oneself, predicts prospective psychotic episodes (over a follow-up period of 5 years).

The study is a longitudinal hypothesis-testing of previous cross-sectional findings Reference Salokangas, Heinimaa, Svirskis, Laine, Huttunen and Ristkari[7], which revealed an association between NAO and escalating levels of psychoticism: broad vulnerability to psychosis, defined by self-reported psychotic-like experiences, and high risk of psychosis, assessed according to international criteria for prodromal/at risk mental states (i.e. transient or attenuated psychotic symptoms; presence of cognitive-perceptual basic symptoms; putative genetic risk and recent functional decline).

The study holds substantial clinical and conceptual appeal for several reasons. First, it shows that, although not formally included among the risk-indicators of transition to psychosis, certain forms of interpersonal sensitivity are actually predictive of imminent risk of psychosis.

Second, it coheres with other empirical studies pointing to certain forms of disturbed intersubjective experience as a feature of schizophrenia-related psychosis Reference Klosterkotter, Ebel, Schultze-Lutter and Steinmeyer[2,Reference Oshima, Okimura, Yukizane, Yasumi, Iwawaki and Nishikawa5]. This includes a cluster of basic symptoms indicative of “interpersonal vulnerability”, such as reduced ability to maintain or initiate social contacts, disturbances of expressive-emotional responsiveness, increased emotional reactivity to routine social interactions and unstable feelings of self-reference Reference Klosterkotter, Ebel, Schultze-Lutter and Steinmeyer[2].

Finally, in our view, the study points to the heuristic value of adopting a sociorelational perspective in the clinical assessment of at-risk mental states. Despite the fact that various social inadequacies are frequently observed in prepsychotic prodromal phases, our current descriptive-conceptual repertoire is relatively modest, particularly within the field of early detection. In this context, the closest criteriological notion is attenuated negative symptoms Reference Lencz, Smith, Auther, Correll and Cornblatt[3], recently re-emphasized within the context of “early and broadly defined psychosis risk mental states” Reference Keshavan, Delisi and Seidman[1].

However, even attenuated negative symptoms, such as initial interpersonal withdrawal or declining social competences, might be of limited clinical specificity if only mapped onto the mere behavioural level (e.g. they can easily be secondary to non-psychotic psychopathology such as anxiety or depressive disorders). But the same manifestations, when considered within a phenomenological context—that is, in relation to subtle feelings of self-reference and an initial paranoid interpersonal sensitivity—might ultimately be more illuminative of impending psychotic risk. Nonetheless, besides the “soft” paranoid tainting of social experiences, other salient experiential changes in the social self are also worth more focused clinical attention. For example, in the early prodrome of schizophrenia, the experience of interpersonal relatedness is often profoundly changed—although not necessarily in a paranoid configuration. This phenomenological alteration may exhibit a number of characteristics: a felt sense of loss of spontaneity (i.e., naturalness) in the very way of relating to and communicating with others; social hypo-hedonia (i.e. reduced vital engagement, social drive, a loss of pleasure in interactions with peers); experienced alterations of intersubjective resonance (e.g. over-impressionability or fading of empathic feelings towards others); propensity to de-socializing and solipsistic modes of experience (e.g. in the sense of alienation from the naturalness of common sense, or extraordinary, quasi-grandiose feelings of superhuman capacities and exceptionality, or radical incommunicability and dehumanization) (Reference Moller and Husby[4] and Reference Stanghellini[8] for clinical examples). In our view, this diversity of experiential phenomena calls for a more systematic clinical focus on the intersubjective manifestations of vulnerability to psychosis, particularly with respect to the phenomenological underpinnings of social interaction.