Ethological Approaches to Psychiatric Disorders: Focus on Depression and Schizophrenia

Abstract

Behavioural observation of psychiatric patient groups using ethological methodology has never been a mainstream approach in psychiatry. In the present review article it is argued that the assessment of non-verbal behaviour in psychiatric disorders has much to offer to clinicians. Based on a Medline survey, the literature on ethological observation of patients with depression and schizophrenia was reviewed. Ethological observation of psychiatric patient groups has the potential to improve prediction of course and outcome of severe psychiatric disorders at a very early stage. Disadvantages of the ethological approach lie within its technical requirements and its inherent time-consuming evaluation of data. A full appreciation of psychiatric disorders may require answers to questions to the proximate causes and evolutionary (Darwinian) trajectories of behaviour, an approach that may well be expanded to maladaptive cognition and emotion.

Keywords

depression ethology prediction of course and outcome schizophrenia

Ethology, as conceptualized in a seminal paper by Nico Tinbergen on the occasion of Konrad Lorenz' 60th birthday, deals with the causation, ontogeny, survival value and evolutionary trajectories of behaviour [1]. The first two of these so-called ‘why’ questions concern the proximate causes of behaviour, which pertain to the processing of internal and external stimuli (physiology), and the ontogenetic development of behaviours over an individual' lifespan. For example, questions related to ontogeny inquire how behaviour changes with ageing, and which early experiences are necessary for a given behaviour to develop properly. The other two questions, referred to as the ultimate causes, address the evolutionary processes that have shaped the behaviour. One question concerns the selective advantages or biological function of behaviour, which includes the analysis of how behaviour impacts on an individual' inclusive fitness (i.e. survival and reproductive success); the other traces the phylogeny of behaviour, which involves cross-species comparison of behaviour, and concerns how the behaviour might have arisen and how it was modified in the course of evolution [1]. Both the proximate and the ultimate causes are regarded as complementary in explanatory power of behaviour.

Traditionally, psychiatry overlaps with ethology regarding the search for proximate causes of behaviour (albeit psychiatry focuses on pathological variants). But even though leading scholars recognized the importance of behavioural observation from the very beginning of psychiatry as a medical discipline in the 19th century [2,3], ethological research based on modern evolutionary conceptualizations has received relatively little attention in the (recent) past. Likewise, psychiatry has largely disregarded the value of considering the evolved function of behaviour such that the messages of non-verbal patterns of behaviour are inadequately understood. Even more so, psychiatry has omitted the possibility that the four questions can be applied to emotional and cognitive processes as well.

A large part of non-verbal behaviour is out of conscious control. Facial expressions of emotion, gestures and body postures are by and large involuntary. As such, the information provided by the analysis of non-verbal behaviour may differ from the information via verbal report [4]. If, for example, a person tries to put on a friendly smile, even if he or she does not feel like that, the smile can easily be uncovered as a ‘fake smile’ by experienced observers, simply because voluntary control over the upper part of the face (eye region) is weak compared to the lower face [5].

The failure of psychiatry to categorize observable behaviours in a framework grounded in human (behavioural) biology has greatly hampered studies into diagnostic accuracy, treatment and prognosis of psychopathological conditions. For example, many experienced psychiatrists are well familiar with a vague feeling that something is wrong in a patient who is developing a psychotic disorder. This intuitive recognition of subtle abnormalities in behaviour, known as ‘praecox feeling’ [6], is widely believed to be a reliable tool in psychiatric examination [7], although it largely lacks scientific appreciation.

In spite of the ignorance of mainstream psychiatry, a handful of psychiatric research groups have built upon the new discipline of human ethology [8], which emerged as an extension of animal ethology in the second half of the 20th century. They basically suggest that ethological research in psychiatric patient groups not only can make a valuable contribution to psychiatry and to insights into psychopathology, but also to the construction of an empirically testable framework for the understanding of abnormal behaviours [9–16]. In the 1950s, for example, Staehelin carried out systematic observation of patients with chronic schizophrenia on the ward and recognized a marked autistic ‘a-sociality’ on the one hand, and the emergence of a strong pecking order or hierarchy on the other [17]. In the late 1950s and early 1960s, ethological research in psychiatric patient groups as well as with regard to a cross-species perspective was further advanced in the pioneering work of Ploog [10,18,19]. Ploog was among the first to highlight the similarities of stereotyped behaviours observed in chronic schizophrenia with instinctual behaviours in animals [18]. In a similar vein, the Tinbergen and Tinbergen analysis of the non-verbal behaviour of autistic children [20] has been a source of inspiration for several other ethological studies into various types of psychopathology, although their approach to autism was later refuted. Finally, the Bowlby attachment theory has become one of the most influential ethological theories in psychiatry [21,22]. One of the core ideas advanced in the Bowlby theoretical model is that early social interaction between mother and newborn plays a crucial role in the social development of the offspring later in life, and that anomalous early attachment experiences may predispose the child to psychopathology later in life. By now, this core idea is generally accepted and utilized in psychological and psychiatric clinical practice.

When looking at systematic behavioural observation in psychiatric patient groups it becomes evident that most studies have concentrated on two phenotypes of psychiatric disorders: depression and schizophrenia. The purpose of the present review, based on a Medline survey, was to summarize the most important findings from ethological observation in depression and schizophrenia. Accordingly, we deliberately excluded the extensive literature on facial expressions of emotion in both depression and schizophrenia, on which excellent reviews exist [23,24]; neither do we discuss in depth the vast literature on the impact of poor attachment on psychological problems. Instead, we will try to put forth the idea – or advocate its revival – that the understanding of psychopathology necessitates a full appreciation of all four questions raised by Tinbergen, that is, both proximate and ultimate factors.

Ethological analysis of behaviour

It is assumed that >60% of human communication is effected via non-verbal behaviour [23,25]. Thus, one can argue that an ethological analysis of psychiatric patients’ non-verbal interactions with others may indeed contribute to a better understanding of the role of interpersonal processes in the onset and course of the disorder.

A basic assumption that underlies the ethological analysis of behaviour is that the behaviour of the species under study is hierarchically organized. Several elements of behaviour may or may not share a common causal factor and a common (biological) function. That is, elements of behaviour that occur in the same context but not in others are assumed to share a similar cause that is related to this context. In addition, behavioural elements that produce a similar effect are assumed to be functionally homologous. Based on contextual evidence and consequential evidence, these elements are pooled into a higher-order factor. In turn, several higher-order factors together can constitute the behavioural repertoire that serves the achievement of a biological motivation (e.g. reproduction, territorial behaviour, social behaviour etc.). In the observation and registration of behaviour ethologists make use of ethograms. Ethograms are catalogues of discrete elements of behaviour that make part of the behavioural repertoire of the species under study. The behavioural elements that constitute an ethogram need to be sufficiently distinguishable from other elements of behaviour (i.e. sufficiently uniform and recurrent) in order to allow the registration of objective parameters (e.g. frequency and duration).

Most ethological studies into psychopathological conditions have used (variants of) an ethogram developed by Grant [26]. This ethogram focuses on the coding of human behaviour during interviews and consists of >100 different codable simple or complex movements such as eye-blink, gaze direction and facial movements, as well as body posture. These behavioural elements are allocated to one of four categories, namely Flight, Assertion, Contact, and Relaxation. For example, drawing back one' mouth corners represents an involuntary and unconscious expression of submission and is often motivated by a tendency to escape a situation. In contrast, an eyebrow flash invites social interaction [8] and is accordingly grouped with other affiliation signals as ‘contact’ behaviour [26,27]. In human interaction, elements that indicate motivational ambivalence, for example between flight and assertion, referred to as ‘displacement behaviours’ [28], occur frequently in clinical interview situations, and these elements can be observed across diagnostic categories. Based on Grant' ethogram and its variants as used by Polsky and McGuire, Schelde et al. and Pedersen et al. [13],[29–31], Troisi et al. have refined and simplified this ethogram in the Ethological Coding System for Interviews (ECSI), [15,32]. It consists of 37 behavioural elements that are grouped according to their meaning. The resulting factors are eye contact, affiliation, submission, pro-social behaviour, flight, assertion, gesture, displacement, and relaxation.

In a similar vein, Bouhuys et al. developed an ethogram to describe the behaviour of depressed patients and their interviewers during a clinical interview. This ethogram is based on findings that single elements of observable behaviour of depressed patients are associated with the patients’ clinical state [33]. In addition, they based their analyses on contextual and consequential evidence that behaviour as displayed during speaking may have a different cause and a different impact on the conversation partner when compared to the same behaviour as displayed while listening (e.g. gaze and gestures). The ethogram consists of 12 elements of behaviour that are described in terms of frequency and duration during speaking and frequency and duration during listening. Based on a factor analysis, the resulting behavioural elements are pooled into six behaviour patient factors called Restlessness-1, Restlessness-2, Active listening, Speech, Eagerness, and Speaking Effort [33]. In a similar approach Bouhuys and van den Hoofdakker also constructed an ethogram for the interviewer' behaviour [34]. This resulted into seven factors that describe the interviewer' behaviour during the clinical interview. Comparable factors with the patients are Restlessness-1, Restlessness-2, Active Listening, and Speech. Specific interviewer factors are Turn Taking, Change Looking, and Encouragement.

Ethology of depression and schizophrenia

Several studies have investigated whether and how ethograms differ between psychiatric patients and healthy controls, and between patients with different types of psychopathology, respectively. McGuire and Polsky investigated whether ethological methods can be used to identify different types of psychopathology [29,35,36]. They observed behaviour of patients with depression (n = 6), schizophreniform disorder (n = 6), and personality disorder (n = 4) in semi-naturalistic settings on the ward. Patients with depression displayed non-social behaviours most frequently and invited others to social interaction less than patients with schizophrenia or personality disorders. They also found that during the time-course of hospitalization the interpersonal space between the patient and others changed in patients with schizophrenia and in patients with personality disorder, but not in depressed patients. Overall, the amount of social behaviour increased with clinical improvement [29]. Differences in observed social behaviour between patients who would improve and those who would not emerged in the second week of hospitalization. Behaviour as observed during the first week, however, did not predict subsequent clinical improvement. Fossi et al. also demonstrated that the non-verbal behaviour of depressed patients (n = 29) differs from that of other non-depressed psychiatric patients (n = 6) [37]. Depressed patients received less verbal behaviour from others and displayed less behaviour that was associated with affiliation than other patients. Furthermore, they replicated the findings of McGuire and Polsky, demonstrating that social behaviour increases when depression improves. Overall, they found that the behaviour of depressed patients normalized with clinical improvement. As in the study by Polsky and McGuire [29], the behavioural profile of patients as assessed at hospital admission was not associated with the subsequent course of depression. Schelde et al. investigated the changes in behaviour of five depressed patients over the course of treatment [31]. They also found that clinical improvement was accompanied by an increase in social behaviour. Schelde replicated these findings in a larger group (n = 11) [38,39]. He suggested that lack of social behaviour could be interpreted as a marker of depression and that increases in social behaviour might be a marker of recovery. Taken together, ethological registrations of behaviour of psychiatric patients as displayed on the wards indicate that the non-verbal behaviour of depressed patients is distinguishable from that of patients with other psychiatric disorders. It appears that, in particular, social behaviour is affected by depression.

In contrast to the semi-naturalistic ward observations, Jones and Pansa observed the behaviour of depressed patients (n = 23), schizophrenia patients (n = 20), and controls (n = 43) during a clinical interview at hospital admission [40]. Compared to healthy controls, depressed patients demonstrated both shorter durations and lower frequencies of smiling and looking at the interviewer. During the early phase of the interview, eye contact was also shorter than in patients with schizophrenia. Furthermore, when compared to patients with schizophrenia as well as to healthy controls, depressed patients demonstrated a shorter duration and lower frequency of body-focused hand movements. With clinical improvement, particularly the frequency and duration of smiles during the first 2 min of the interview increased. Jones and Pansa furthermore found that the time-course of the behaviour over the interview differed between depressed patients and those with schizophrenia [40]. Hence, in line with findings from ward observations, the findings from clinical interviews also point out that social behaviour distinguishes depressed patients from patients with schizophrenia and from healthy controls.

Troisi et al. demonstrated that the behaviour of medication-free young Italian male patients with schizophrenia as displayed during an interview differed from that of healthy controls [32]. Compared to the controls, patients demonstrated less pro-social behaviour, less gestures and less conflict behaviour (displacement activity) than controls did. In a gender-mixed group of 44 German patients with schizophrenia (n = 34) and schizoaffective disorder (n = 9) who were on second-generation antipsychotics, Brüne et al. found that patients displayed less relaxation, less flight, less pro-social behaviour and less affiliation than healthy controls [41]. They argued that the differences between the two studies could perhaps reflect subtle cultural differences and gender differences, as well as differences according to the interviewer' gender [41]. The finding, however, that pro-social behaviour encouraging social interaction has consistently been shown to distinguish best between patients with schizophrenia and controls of both sexes, across cultures, and perhaps relatively independent of the interviewer' gender, could buttress the observation of early cross-cultural psychiatrists that patients with schizophrenia are cross-culturally much more similar in behaviour than the healthy individuals of the respective cultures [42].

Prediction of course and outcome in depression and schizophrenia based on non-verbal behaviour

The findings described in the previous section point to an association between disturbed non-verbal interpersonal behaviour and depression and schizophrenia. Further evidence that these disturbed non-verbal interactions play a causal role in depression and schizophrenia comes from studies that investigated whether disturbed non-verbal interactions can also predict the course of depression and schizophrenia. Ranelli and Miller investigated non-verbal behaviour of 18 depressed patients in relation to the course of depression over a 6 week follow up [43]. They found that patients who would not improve displayed a high frequency of body-focused adaptors, posture shifts and speech pauses. Patients who would improve displayed long speech pauses and head aversions. Using the ECSI, Troisi et al. demonstrated that non-response to a 5 week treatment with amitriptyline was predicted by high levels of submissive behaviour, affiliation, and assertive behaviour (n = 22) [44]. Bouhuys and Albersnagel investigated the predictive quality of their ethogram with respect to the course of depression over a 10 week follow up in a group of 31 hospitalized major depressed patients. They found that high levels of the patients’ Restlessness-1 and Speaking Effort and low levels of Active Listening were associated with a poor outcome of depression [45]. Moreover, it was found that the observable behaviour of the interviewer was also associated with the subsequent course of depression: when compared to patients with a favourable course of depression, interviewers displayed relatively lower levels of Restlessness-1 and Active Listening and relatively higher levels of Encouragement toward patients who would show a poor outcome [34]. Geerts et al. were able to replicate these findings in a group of 24 outpatients with seasonal affective disorder (SAD) [46]. The findings in patients with SAD resembled those in patients with major depression mostly with respect to high levels of patients’ Speaking Effort and of interviewers’ Encouragement as predictors of poor outcome of depression. In contrast to patients in depressed states, Bos et al. found in 51 remitted depressed patients that low levels of Speaking Effort predicted recurrence of depression during a 6 month follow up [47].

Hale et al. used the ethogram developed by Bouhuys et al. to investigate the interaction between 26 depressed patients and their partners and between patients and an unfamiliar person [48]. That study demonstrated that partners of those patients who would not improve during the follow-up period displayed higher levels of Speech than those of patients who would improve. Neither the behaviour of the patients in these interactions, nor that of unfamiliar people was associated with the subsequent course of the depression.

Geerts et al. investigated the nature of the interrelationship between the patients’ Speaking Effort and their interviewers’ Encouragement [49]. For this purpose they interviewed 11 mildly depressed patients. During the interview the interviewer experimentally controlled the levels of displayed Encouragement. The patients received 10 min with high levels of Encouragement and 10 min with low levels of Encouragement in a cross-over design. As hypothesized, the patients displayed higher levels of Speaking Effort during the high Encouragement condition than during the low Encouragement condition. These findings indicate that while interacting, patients match the amount of their displayed Speaking Effort to the amount of Encouragement as displayed by the interviewer. In a subsequent study, Geerts et al. (unpubl. data) used a pseudo-interaction paradigm to statistically confirm this causal interrelatedness in naturalistic clinical interviews in a group of 31 patients with major depression and in a group of 80 patients with SAD [50,51]. They demonstrated that the levels of the patients’ Speaking Effort and those of the interviewers’ Encouragement in the interviews were significantly more similar than may be expected by chance.

In healthy subjects, non-verbal convergence between conversation partners during an interaction underlies satisfaction with the interaction and mutual attraction [52,53]. In remitted depressed patients Geerts et al. also found that the more patients and interviewers matched the levels of their non-verbal displays of Speaking Effort and Encouragement to a similar level, the more satisfied patients turned out to be with the interview [54]. Tickle-Degnen and Rosenthal demonstrated that non-verbal convergence in clinical interactions underlies rapport [55]. Geerts hypothesized that lack of non-verbal convergence may be a mechanism that is involved in the onset and course of depression [56]. This hypothesis has been tested in five independent groups of depressed patients so far (total n = 245) [57–60; Geerts et al., unpubl. obs.;]. In all but one study it was found that lack of non-verbal convergence between the patients’ Speaking Effort and the interviewers’ Encouragement predicted an unfavourable short-term outcome of depression. Moreover, in a group of 102 remitted depressed patients the lack of non-verbal convergence between patients and an interviewer was shown to predict depression-relapse within a 2 year follow up [54,61]. In that same patient sample Bos et al. found that patients who reached low levels of non-verbal convergence with their interviewer were at higher risk of being involved in negative interpersonal events during the follow up [62]. They furthermore demonstrated that these negative events mediated non-verbal convergence to recurrence of depression.

Geerts and Bouhuys re-analysed the Hale et al. data to investigate whether the course of depression could also be predicted by the non-verbal convergence of depressed patients and their partner and between depressed patients and unfamiliar persons [48; Geerts et al., unpubl. obs.;]. Again, a pseudo-interaction paradigm was applied to demonstrate that non-verbal convergence between the patients’ Speaking Effort and the conversation partners’ Encouragement actually occurred [50,51]. They found that non-verbal convergence between the patients’ Speaking Effort and Encouragement as displayed by unfamiliar persons occurred in the interactions of patients who would improve during the follow up. Non-verbal convergence did not occur, however, in the interactions of patients who would not improve. In the interactions between patients and their partners similar findings were obtained. The patients’ Speaking Effort and the partners’ Encouragement were causally interrelated in the interactions of patients who would improve. In the interactions of patients who would not improve, however, this causal interrelatedness did not occur. Based on these findings one can conclude that the results on the causal interrelationship between patients’ Speaking Effort and Encouragement as displayed by clinical interviewers can be generalized to other social interactions of patients. Moreover, the findings demonstrate that non-verbal convergence during interactions with other people is also associated with the subsequent course of depression. Taken together, these findings favour the hypothesis that lack of non-verbal convergence between patients and conversation partners is a mechanism that underlies lack of satisfaction with these interactions and negative interpersonal events that in turn provoke (an unfavourable course of) depression. It has been demonstrated that lack of non-verbal convergence can neither be explained by the patients’ personality traits Neuroticism and Extraversion nor by the patients’ cognitive interpretation of facially expressed emotions [58,59,62]. Hence, non-verbal communication between depression-prone people and people from their social environment on the one hand, and personality traits and social cognition on the other, appear to play independent roles in the onset and course of depression.

To summarize, behavioural observations of depressed patients during a clinical interview confirm findings from ward observation that impaired social behaviour is involved in depression. Moreover, in contrast to the ward observations, the non-verbal behaviour as registered from clinical interviews does predict the subsequent course of depression. This predictive quality is not restricted to the patients’ behaviour. Also, the behaviour of the conversation partner is associated with the subsequent course of depression. Ward observations have shown that depressed patients engage less often in social interactions. The observations of clinical interviews can be interpreted in terms of a measure of the quality of the interaction. Hence, a possible explanation of the difference in predictive quality between ward observations and those from clinical interviews is that the quality of social interactions plays a causal role in the onset and course of depression, more than the quantity of social interactions. This explanation is supported by findings that the interaction between the behaviour of patients and conversation partners is associated with the subsequent course of depression. Alternative explanations, however, are also possible. The number of patients who participated in the clinical interviews exceeds that of patients who were studies in the ward observations. Hence, the lack of predictive quality of ward observations may be a simple problem of lack of power. In addition, the circumstances of clinical interviews are more standardized than those of ward observations. This may also explain why behaviour as observed during clinical interviews does predict the course of depression whereas behaviour observed of patients on the ward does not.

In contrast to the number of studies on the predictive quality of observable behaviour with respect to the onset and course of depression, little is known on this predictive quality in patients with schizophrenia. Troisi et al. found in 18 patients with schizophreniform disorder that patients with a good prognosis demonstrated more eye contact with an interviewer than those with a poor prognosis [63]. Troisi and Moles interpreted this finding as a behavioural correlate of poor rapport [64]. Hence, one can suggest that, as in depression, poor non-verbal interaction is associated with poor outcome. To the best of our knowledge, no other studies exist that have investigated whether observable behaviour of either patients or interviewers can predict the course of schizophrenia and related disorders in a prospective design. Walker et al. and Schiffman et al., however, demonstrated that already, at an early age, the social behaviour of children who turned out to develop schizophrenia later in life differs significantly from those who turned out to develop no psychiatric disorder [65,66]. We are not aware of studies that demonstrate such differences between the social behaviour of depression-prone children and those who are not prone to psychiatric disorder.

Discussion

In this article we have reviewed ethological naturalistic observational studies of depression and schizophrenia. We argue that Tinbergen' four questions that address the causes of behaviour at the proximate and the ultimate level [1] can be a useful and informative tool for behavioural observations in psychiatric patient groups and are a necessary tool for a full and comprehensive understanding of psychopathology in an evolutionary perspective, for several reasons. First, ethology helps categorize behaviours in terms of their communicative meaning. Drawing on cross-species as well as cross-cultural issues, behaviours observed in psychiatric patients can be put in an empirically testable framework. For example, behaviours seen in depression or schizophrenia such as crouching postures, averted gaze or displacement activities often reflect defensive strategies [67,68]. Accordingly, changing patterns of behaviour, for example observed in a patient who is able to engage less often in such defensive positions, can be linked to questions about clinical improvement, even before the patient (or clinician) becomes subjectively aware of it. More importantly, increase of displacement activities, if recognized, can alert clinicians to check for clinical deterioration. For example, such behaviour may indicate increasing motivational conflict and ambivalence, which can be a sign of impending suicidal behaviour [31]. These examples of behavioural analyses based on ethological methodology explicitly assume that behaviours found in clinical conditions are not qualitatively distinct from behaviours in healthy individuals but are different by degree, that is, intensity, frequency or contextual inappropriateness [41,69]. Second, behavioural observation is often much more reliable than subjective report, because it is much less under conscious control compared to verbal communication such that an individual' real motives cannot so easily be concealed [32]. For the very same reason, an ethological approach is at least as valuable as (evolutionary) psychological approaches based on questionnaires [70]. The studies cited here clearly demonstrate the usefulness of ethological observations in revealing the mechanisms that underlie the onset and course of disorders. Third, standard rating scales utilized in clinical assessments critically depend on the clinician' impression of patients’ non-verbal behaviour. Even though manuals available for clinical rating scales hardly ever recur on ethological theory, clinicians intuitively use their species-specific endowments for deciphering non-verbal expressions in therapist–patient interactions. The extent to which actual clinical judgements rely on unconsciously perceived communicative signals sent by patients compared to their subjective report is a highly underresearched topic in clinical psychiatry [41].

Ethological observation in psychiatric patient groups also has, however, a number of disadvantages. Ethological methodology is still outside psychiatric mainstream research, because it is extremely time-consuming, and because it requires training that medical or psychology students usually do not receive (because ethology is not part of the curriculum), thus, ethological terminology is unfamiliar to most clinicians. Moreover, behavioural observation in psychiatric patient groups has very rarely been linked with physiological measures such as neurotransmitter activity or genetic variation. These disadvantages ought to be overcome if psychiatry wants to survive as a medical discipline rooted in the natural sciences.

Apart from psychiatry, we also believe that ethology has much to offer to evolutionary psychology. As we have argued, an evolutionary explanation of psychopathology makes no sense if it does not fit with the empirically based findings on causation and ontogeny. Ethological studies may provide empirical data that can be used to test evolutionary explanations. For instance, the results of ethological observations together with findings from other studies that point to a decreased fitness in depression and findings that even mild signals of depression induce rejection in others [71] challenge evolutionary explanations that depression may be explained as a strategy to seek support and to decrease social threat [72]. In fact, empirical evidence exists that the non-verbal social behaviour of depression-prone people may generate the negative interpersonal events that in turn provoke depression [61]. We therefore assert that a comprehensive evolutionary approach of psychopathology requires the full exploration of all possible explanations of how these disorders may have been maintained in the human population over the course of evolution.

The study of psychiatric patient groups, in comparison to healthy controls, for instance, can identify important details of proximate and ultimate factors of human life. This way, ethology, sociobiology, and evolutionary psychology may return to the methodology of Darwin' ‘On the Expressions of Emotions in Man and Animals’, first published in 1872 [73]. Remarkably, Darwin drew heavily on facial expressions observed in patients with mental illnesses and included several pictures taken by Crichton-Browne, who later became one of the leading psychiatric authorities in Great Britain.

Focusing on non-verbal behaviour in psychiatric patient groups does by no means suggest that patients’ verbal report and subjective experiences are less important sources of information. Non-verbal behaviour, however, has grossly been neglected in psychiatric research in the past, compared to the subjective side of mental illness.

In any event, cognitive and emotional processes have an evolutionary history in essentially the same way that behaviour has. Accordingly, they should be analysed in exactly the same way as behaviour. It could therefore be a fruitful approach to combine research into cognitive deficits with that on non-verbal behaviour, as has recently been demonstrated by Brüne et al. [74]. They found that patients with schizophrenia who understand less well what is going on in other people' minds (i.e. impaired ‘theory of mind’ or ‘mentalizing’) use expressive facial expressions less often than patients with preserved mentalizing abilities [74]. Interactions between cognition and behaviour should, therefore, be more often empirically studied than is currently the case. The future of psychiatry, at least in our opinion, will be to find answers to all four questions proposed by Tinbergen [1] with regards to behaviour, emotion, and cognition.

References

1. Tinbergen

. On aims and methods of ethology. Zeitschr Tierpsychol 1963;20:410–433.

2. Kahlbaum

. Die Katatonie oder das Spannungsirresein. Eine klinische Form psychischer Krankheit., Berlin: Hirschwald, 1874/1973 (English translation: Catatonia: the tension insanity).

3. Brüne

. Rethinking Karl Ludwig Kahlbaum' contribution to biological psychiatry. Aust N Z J Psychiatry 2000;34:873–874.

4. Gotlib

Robinson

. Responses to depressed individuals: discrepancies between self-report and observer-rated behavior. J Abnorm Psychol 1982;91:231–240.

5. Ekman

. Emotions revealed: recognizing faces and feelings to improve communication and emotional life. New York: Times Books/Henry Holt, 2003.

6. Rümke

. Het kernsymptoom der schizophrenie en het ‘praecoxgevoel’. Tijdschr Geneeskunde 1941;81:4516–4521.

7. Sagi

Schwartz

. The ‘praecox feeling’ in the diagnosis of schizophrenia; a survey of Manhattan psychiatrists. Schizophr Res 1989;2:35.

8. Eibl-Eibesfeldt

. Human ethology. Hawthorne, NY: Aldine de Gruyter, 1989.

9. Lorenz

. Über die Bildung des Instinktbegriffs. Naturwissenschaften 1937;25:289–300.

10.

10. Ploog

. Verhaltensforschung und Psychiatrie. Gruhle

Jung

Mayer-Gross

Müller

. Psychiatrie der Gegenwart, Bd. I/1B., 1. Aufl. Berlin: Springer, 1964;:291–443.

11.

11. Tinbergen

. Ethology and stress diseases. Science 1974;185:20–27.

12.

12. McGuire

Essock-Vitale

Polsky

. Psychiatric disorders in the context of evolutionary biology: an ethological model for behavioural changes associated with psychiatric disorders. J Nerv Ment Dis 1981;169:687–704.

13.

13. Schelde

. Ethological research in psychiatry. Ethol Sociobiol 1994;15:349–368.

14.

14. Dixon

. Ethological strategies for defence in animals and humans: their role in some psychiatric disorders. Br J Med Psychol 1998;71:417–445.

15.

15. Troisi

. Ethological research in clinical psychiatry: the study of nonverbal behavior during interviews. Neurosci Biobehav Rev 1999;23:905–913.

16.

16. Bouhuys

. Ethology and depression. Philippot

Feldman

Coats

. Nonverbal behavior in clinical settings. New York: Oxford University Press, 2003; 233–262.

17.

17. Staehelin

. Gesetzmäßigkeiten im Gemeinschaftsleben schwer Geisteskranker. Schweiz Arch Neurol Psychiatr 1953;72:277–298.

18.

18. Ploog

. Endogene Psychosen und Instinktverhalten. Fortschr Neurol Psychiatr 1958;26:83–98.

19.

19. Ploog

. Experimentelle Verhaltensforschung. Nervenarzt 1966;37:443–447.

20.

20. Tinbergen

Tinbergen

. Early childhood autism: an ethological approach. Advances in Ethology 10. Berlin: Parey, 1972.

21.

21. Bowlby

. Attachment and loss. Volume 1: attachment. New York: Basic Books, 1969.

22.

22. Bowlby

. Attachment and loss. Volume 3: loss: sadness and depression. Middlesex: Penguin, 1981.

23.

23. Mandal

Pandey

Prasad

. Facial expressions of emotions and schizophrenia: a review. Schizophr Bull 1998;24:399–412.

24.

24. Gaebel

Wölwer

. Facial expressivity in the course of schizophrenia and depression. Eur Arch Psychiatry Clin Neurosci 2004;254:335–342.

25.

25. Cahn

Frey

. Listeners’ perceived verbal and nonverbal behaviors associated with communicators’ perceived understanding and misunderstanding. Percept Motor Skills 1992;74:1059–1064.

26.

26. Grant

. An ethological description of non-verbal behaviour during interviews. Br J Med Psychol 1968;41:177–183.

27.

27. Grant

. Human facial expression. Man 1969;4:525–536.

28.

28. Tinbergen

. ‘Derived’ activities; their causation, biological significance, origin, and emancipation during evolution. Q Rev Biol 1952;27:1–32.

29.

29. Polsky

McGuire

. Observational assessment of behavioral changes accompanying clinical improvement in hospitalized psychiatric patients. J Behav Assess 1980;2:207–223.

30.

30. Pedersen

Schelde

JTM

Hannibal

Behnke

Nielsen

Hertz

. An ethological description of depression. Acta Psychiatr Scand 1988;78:320–330.

31.

31. Schelde

JTM

Pedersen

Hannibal

Behnke

Nielsen

Hertz

. An ethological analysis of depression: comparison between ethological ratings and Hamilton ratings of five endogenously depressed patients. Acta Psychiatr Scand 1988;78:331–340.

32.

32. Troisi

Spalletta

Pasini

. Non-verbal behaviour deficits in schizophrenia: an ethological study of drug-free patients. Acta Psychiatr Scand 1998;97:109–115.

33.

33. Bouhuys

Jansen

van den Hoofdakker

. Analysis of observed behaviors displayed by depressed patients during a clinical interview: relationships between behavioral factors and clinical concepts of activation. J Affect Disord 1991;21:79–88.

34.

34. Bouhuys

van den Hoofdakker

. A longitudinal study of interaction patterns of a psychiatrist and severely depressed patients based on observed behavior: an ethological approach of interpersonal theories of depression. J Affect Disord 1993;27:87–99.

35.

35. McGuire

Polsky

. Behavioural changes in hospitalized acute schizophrenics. J Nerv Ment Dis 1979;167:651–657.

36.

36. McGuire

Polsky

. Ethological assessment of stable and labile social behaviors during acute psychiatric disorders: clinical applications. Psychiatry Res 1980;3:291–306.

37.

37. Fossi

Faravelli

Paoli

. The ethological approach to the assessment of depressive disorders. J Nerv Ment Dis 1984;172:332–341.

38.

38. Schelde

. Major depression: behavioural markers of depression and recovery. J Nerv Ment Dis 1998;186:133–140.

39.

39. Schelde

. Major depression: behavioral parameters of depression and recovery. J Nerv Ment Dis 1998;186:141–149.

40.

40. Jones

Pansa

. Some nonverbal aspects of depression and schizophrenia occurring during the interview. J Nerv Ment Dis 1979;167:402–409.

41.

41. Brüne

Sonntag

Abdel-Hamid

Lehmkämper

Juckel

Troisi

. Non-verbal behavior during standardized interviews in patients with schizophrenia spectrum disorders. J Nerv Ment Dis 2008;196:282–288.

42.

42. Pfeiffer

. Transkulturelle Psychiatrie 2nd edn. Stuttgart: Thieme, 1994.

43.

43. Ranelli

Miller

. Behavioral predictors of amitriptyline response in depression. Am J Psychiatry 1981;138:30–34.

44.

44. Troisi

Pasini

Bersani

Grispini

Ciani

. Ethological predictors of amitriptyline response in depressed outpatients. J Affect Disord 1989;17:129–136.

45.

45. Bouhuys

Albersnagel

. Do interactional capacities based on observed behaviour interfere with improvement in severely depressed patients?. J Affect Disord 1992;25:107–116.

46.

46. Geerts

Bouhuys

Meesters

Jansen

. Observed behavior of patients with seasonal affective disorder and an interviewer predicts response to light treatment. Psychiatry Res 1995;57:223–230.

47.

47. Bos

Geerts

Bouhuys

. Nonverbal interaction involvement as an indicator of prognosis in remitted depressed subjects. Psychiatry Res 2002;113:269–277.

48.

48. Hale

Jansen

JHC

Bouhuys

Jenner

van den Hoofdakker

. Nonverbal behavioral interactions of depressed patients with partners and strangers: the role of behavioral social support and involvement in depression persistence. J Affect Disord 1997;44:111–122.

49.

49. Geerts

Bouhuys

Bloem

. Nonverbal support giving induces nonverbal support seeking in depressed patients. J Clin Psychol 1997;53:35–39.

50.

50. Bernieri

. Coordinated movement and rapport in teacher–student interactions. J Nonverb Behav 1988;12:120–138.

51.

51. Bernieri

Rosenthal

. Interpersonal coordination: behavior matching and interactional synchrony. Feldman

Rimé

. Fundamentals of behaviour. Cambridge: Cambridge University Press, 1991;:401–432.

52.

52. Cappella

Palmer

. Attitude similarity, relational history, and attraction: the mediating effects of kinesic and vocal behaviors. Commun Monogr 1990;57:161–183.

53.

53. Cappella

Palmer

. The effect of partners’ conversation on the association between attitude similarity and attraction. Commun Monogr 1992;59:180–189.

54.

54. Geerts

van Os

TWDP

Ormel

Bouhuys

. Nonverbal behavioral similarity between patients with depression in remission and interviewers in relation to satisfaction and recurrence of depression. Depress Anxiety 2006;23:200–209.

55.

55. Tickle-Degnen

Rosenthal

. The nature of rapport and its nonverbal correlates. Psychol Inquiry 1990;1:285–293.

56.

56. Geerts

. An ethological approach of interpersonal theories of depression. Thesis. Groningen, Germany: University of Groningen, 1997.

57.

57. Geerts

Bouhuys

van den Hoofdakker

. Nonverbal attunement between depressed patients and an interviewer predicts subsequent improvement. J Affect Disord 1996;40:15–21.

58.

58. Geerts

Kouwert

Bouhuys

Meesters

Jansen

. Nonverbal interpersonal attunement and extravert personality predict outcome of light treatment in seasonal affective disorder. J Affect Disord 2000;59:193–204.

59.

59. Geerts

Bouhuys

. Multilevel prediction of short-term outcome of depression: the role of nonverbal interpersonal and cognitive processes and of personality traits. Psychiatry Res 1998;57:185–191.

60.

60. Geerts

van Os

Gerlsma

. Nonverbal communication sets the conditions for the relationship between parental bonding and the short-term treatment response in depression. Psychiatry Res 2009;165:120–127.

61.

61. Bos

Bouhuys

Geerts

van Os

TWDP

Ormel

. Stressful life events as a link between problems in nonverbal communication and recurrence of depression. J Affect Disord 2007;97:161–169.

62.

62. Bos

Bouhuys

Geerts

van Os

TWDP

Ormel

. Lack of association between conversation partners’ nonverbal behavior predicts recurrence of depression, independently of personality. Psychiatry Res 2006;142:79–88.

63.

63. Troisi

Pasini

Bersani

Di Mauro

Ciani

. Negative symptoms and visual behavior in DSM-III-R prognostic subtypes of schizophreniform disorder. Acta Psychiatr Scand 1991;83:391–394.

64.

64. Troisi

Moles

. Gender differences in depression: an ethological study of nonverbal behavior during interviews. J Psychiatry Res 1999;33:243–250.

65.

65. Walker

Grimes

Davis

Smith

. Childhood precursors of schizophrenia: facial expressions of emotions. Am J Psychiatry 1993;150:1654–1660.

66.

66. Schiffman

Walker

Ekstrom

Schulsinger

Sorensen

Mednick

. Childhood videotaped social and neuromotor precursors of schizophrenia: a prospective investigation. Am J Psychiatry 2004;161:2021–2027.

67.

67. Troisi

. Displacement activities as a behavioral measure of stress in nonhuman primates and human subjects. Stress 2002;5:47–54.

68.

68. Gilbert

. Social mentalities. A biopsychosocial and evolutionary approach to social relationships. Baldwin

. Interpersonal cognition. New York: Guilford Press, 2005; 299–334.

69.

69. Brüne

. Ethological remarks on mannerisms. Conceptualisation and proposal for a definition. Psychopathology 1998;31:188–196.

70.

70. Daly

Wilson

. Human evolutionary psychology and animal behaviour. Anim Behav 1999;57:509–519.

71.

71. Segrin

Dillard

. The interactional theory of depression: a meta-analysis of the research literature. J Soc Clin Psychol 1992;11:43–70.

72.

72. Allen

Badcock

PBT

. Darwinian models of depression: a review of evolutionary accounts of mood and mood disorders. Progr Neuropsychopharmacol Biol Psychiatry 2006;30:815–826.

73.

73. Darwin

. On the expressions of emotions in man and animals. London: John Murray, 1872.

74.

74. Brüne

Abdel-Hamid

Sonntag

Lehmkämper

Langdon

. Linking social cognition with social interaction: non-verbal expressivity, social competence and ‘mentalising’ in patients with schizophrenia spectrum disorders. Behav Brain Funct 2009;5:e6.