استنشاق اکسی توسین داخل بینی پردازش از چهره در پروزوپاگنوزیا تکاملی را بهبود می بخشد

Abstract Developmental prosopagnosia (DP) is characterised by a severe lifelong impairment in face recognition. In recent years it has become clear that DP affects a substantial number of people, yet little work has attempted to improve face processing in these individuals. Intriguingly, recent evidence suggests that intranasal inhalation of the hormone oxytocin can improve face processing in unimpaired participants, and we investigated whether similar findings might be noted in DP. Ten adults with DP and 10 matched controls were tested using a randomized placebo-controlled double-blind within-subject experimental design (AB-BA). Each participant took part in two testing sessions separated by a 14–25 day interval. In each session, participants inhaled 24 IU of oxytocin or placebo spray, followed by a 45 min resting period to allow central oxytocin levels to plateau. Participants then completed two face processing tests: one assessing memory for a set of newly encoded faces, and one measuring the ability to match simultaneously presented faces according to identity. Participants completed the Multidimensional Mood Questionnaire (MMQ) at three points in each testing session to assess the possible mood-altering effects of oxytocin and to control for attention and wakefulness. Statistical comparisons revealed an improvement for DP but not control participants on both tests in the oxytocin condition, and analysis of scores on the MMQ indicated that the effect cannot be attributed to changes in mood, attention or wakefulness. This investigation provides the first evidence that oxytocin can improve face processing in DP, and the potential neural underpinnings of the findings are discussed alongside their implications for the treatment of face processing disorders.

Introduction Face recognition is an important cognitive skill that most people take for granted, yet it depends on a complex set of cognitive and neural processes (Bruce and Young, 1986 and Haxby et al., 2000). In some individuals this process can be selectively disrupted, resulting in a condition termed “prosopagnosia” or “face-blindness”. While prosopagnosia can be acquired following brain injury (e.g., Damasio, Damasio, & Van Hoesen, 1982), many more individuals simply fail to develop normal face recognition abilities (e.g., Bate et al., 2009, Bate et al., 2008 and Behrmann and Avidan, 2005; Bentin, Deouell, & Soroker, 1999; Duchaine et al., 2007, Duchaine and Nakayama, 2006, Jones and Tranel, 2001 and Schmalzl et al., 2008). The latter form of the disorder has been termed ‘developmental prosopagnosia’ (DP; but for a discussion of terminology see Susilo & Duchaine, 2013), and has been attributed to a failure to develop the visual recognition mechanisms necessary for successful face recognition, despite intact low-level visual and intellectual functions. Interestingly, there also appears to be a genetic component to the disorder in at least some individuals (Duchaine et al., 2007 and Grueter et al., 2007). In the last decade it has become increasingly clear that DP represents a significant clinical disorder, with recent reports suggesting that two percent of the population have the condition (Bowles et al., 2009 and Kennerknecht et al., 2006). Although many studies have investigated the cognitive, neural and genetic basis of DP, little attention has been directed towards improving face recognition in these individuals. While some researchers have attempted to remedy face processing deficits using extensive visual training programmes (e.g., DeGutis et al., 2007 and Schmalzl et al., 2008), recent evidence suggests that an alternative methodology warrants investigation. Specifically, in some circumstances, intranasal inhalation of the hormone oxytocin has been found to improve face processing in both healthy participants (e.g., Rimmele et al., 2009 and Savaskan et al., 2008) and individuals with autism (Andari et al., 2010). Oxytocin is a nonapeptide centrally involved in the regulation of basic social and reproductive behaviours, such as cohabitation, gestation, and breastfeeding. It has been found to be crucial for social recognition, grooming, approach behaviour, sexual activity and stress regulation in non-human mammals (e.g., Carter, 1998, Ferguson et al., 2001 and Lim and Young, 2006). Recent evidence demonstrates that oxytocin also facilitates social cognition and pro-social behaviour in humans (Baumgartner et al., 2008, Heinrichs et al., 2009, Mikolajczak et al., 2010 and Zak et al., 2007). Indeed, studies using healthy participants have shown that intranasal inhalation of oxytocin can strengthen memory for social but not non-social stimuli (Guastella, Mitchell, & Dadds, 2008), including faces (Rimmele et al., 2009 and Savaskan et al., 2008). However, the precise influence of oxytocin on face memory remains unclear, as the hormone seems to only improve the recognition of faces displaying particular emotional expressions, and existing studies have reported conflicting findings. For instance, while Rimmele et al. (2009) found oxytocin improved memory for faces displaying both positive and negative expressions, Guastella et al. (2008) observed improved memory for happy but not angry or neutral faces, and Savaskan et al. (2008) reported improved recognition of neutral and angry but not happy faces. While the precise influence of oxytocin on face memory remains to be unravelled, it is pertinent that the hormone has also been found to influence processing strategy. Indeed, oxytocin has been reported to increase the time spent looking at the eye region of the face (Guastella et al., 2008), an area thought to provide critical information for identification (Ellis et al., 1979 and Young et al., 1986). It is of note that this shift in processing strategy has also been reported in individuals with autistic spectrum disorder (Andari et al., 2010), who commonly experience face recognition deficits (Schultz, 2005). The findings discussed above suggest that intranasal inhalation of oxytocin may also facilitate face recognition in DP. The current investigation set out to address this issue, investigating whether oxytocin can improve performance in 10 DPs and 10 matched control participants on a task that assesses the encoding and recognition of new faces. In addition, we also assessed performance on a face matching task that assesses the ability to perceive facial identity (thereby placing minimal demands on long-term memory for faces). This issue is particularly relevant to the current study given that some prosopagnosics also have face perception deficits, and sequential models of face processing predict that such impairments inevitably bring about recognition deficits (e.g., Bruce & Young, 1986). This latter task also represents a novel contribution to the literature, given that no studies have examined the influence of oxytocin on face perception skills.

3. Results 3.1. Adverse side effects and MMQ Adverse side effects were only reported by one DP participant following inhalation of either spray. Specifically, this individual reported a slight headache immediately after oxytocin inhalation, but this had disappeared by the 24-h follow-up. A mixed factorial MANOVA revealed no main effect of spray or group, F(3,16) = .569, p = .643, ƞp2 = .096 and F(3,16) = 1.597, p = .229, ƞp2 = .230, although the interaction between the two factors approached significance, F(3,16) = 2.904, p = .067, ƞp2 = .353. This latter effect was driven by the control (but not DP) participants feeling ‘better’ according to the good–bad scale in the oxytocin condition, while the DP but not the control participants felt calmer in the oxytocin condition. There was a main effect of time, F(6,13) = 4.271, p = .014, ƞp2 = .663, and further analyses revealed this was driven by an increase in wakefulness after the 45 min resting period, F(2,36) = 4.626, p = .016, ƞp2 = .204. However, this effect did not interact with participant group or spray, F(2,36) = 1.914, p = .162, ƞp2 = .096 and F(2,36) = .225, p = .800, ƞp2 = .012. 3.2. Face processing tests A mixed factorial MANOVA revealed a significant improvement in the oxytocin compared to the placebo condition, F(2,16) = 5.944, p = .012, ƞp2 = .426. Univariate tests confirmed that performance was better in the oxytocin rather than the placebo condition on both the CFMT (oxytocin: M = 56.89, SE = 1.79; placebo: M = 53.93, SE = 2.14) and matching test (oxytocin: M = 33.11, SE = .76; placebo: M = 31.10, SE = .96), F(1,17) = 4.975, p = .039, ƞp2 = .226 and F(1,17) = 5.786, p = .028, ƞp2 = .254. While there was no main effect of group on the multivariate analysis, F(2,16) = 2.307, p = .132, ƞp2 = .224, group and spray did interact, F(2,16) = 4.422, p = .030, ƞp2 = .356. Univariate analyses confirmed that this interaction was driven by a greater improvement on the CFMT in DPs compared to controls, but the same effect was not significant for the matching test, F(1,17) = 6.035, p = .025, ƞp2 = .262 and F(1,17) = 2.098, p = .166, ƞp2 = .110 (see Fig. 2). Follow-up comparisons found that while the performance of DPs improved in the oxytocin condition for both the CFMT and matching test, F(1,8) = 7.667, p = .024, ƞp2 = .489 and F(1,9) = 9.238, p = .014, ƞp2 = .507, the same pattern was not observed in control participants, F(1,9) = .040, p = .847, ƞp2 = .004 and F(1,9) = .482, p = .505, ƞp2 = .051. Performance of DP and control participants under placebo and oxytocin conditions ... Fig. 2. Performance of DP and control participants under placebo and oxytocin conditions in (A) the CFMT (maximum score of 72) and (B) the face matching test (maximum score of 40). Asterisks indicate significant between- and within-group differences. Figure options Further analyses focused on the performance of the DP group. While, DP performance in the CFMT placebo condition was significantly lower than that of controls, F(1,17) = 6.308, p = .025, ƞp2 = .262, their performance in this condition was greater than their performance on the original diagnostic version of the CFMT (i.e., the version completed within the diagnostic session: see Table 1), F(1,8) = 8.228, p = .021, ƞp2 = .507. It is possible that this difference may be attributed to a placebo effect, but potential practice effects and the fundamental differences between the stimuli in the experimental CFMTs compared to the original version make this observation tentative. Strikingly, DP performance in the oxytocin condition did not differ from that of controls, F(1,18) = 1.257, p = .277, ƞp2 = .065, and a similar pattern was observed in the matching test. Indeed, while DP performance in the placebo condition remained lower than control placebo performance, F(1,18) = 6.322, p = .022, ƞp2 = .260, DP performance in the oxytocin condition did not differ from control oxytocin scores, F(1,18) = 2.266, p = .150, ƞp2 = .112. A final set of analyses investigated whether the severity of each individual's prosopagnosia predicted the extent of their improvement in the oxytocin condition. Performance on the original version of the CFMT (from the diagnostic session) did not correlate with the extent of the improvement in the experimental CFMT, r = .352, n = 9, p = .353. Likewise, performance on the CFPT (a face perception test from the diagnostic session) did not correlate with the extent of improvement on the matching test, r = .073, n = 10, p = .842 (see Fig. 3). (A) Scores on the original version of the CFMT plotted against the level of ... Fig. 3. (A) Scores on the original version of the CFMT plotted against the level of improvement in the oxytocin condition on the new versions of the CFMT (i.e., the difference between the scores achieved in the oxytocin and placebo conditions), and (B) scores on the CFPT plotted against the level of improvement in the oxytocin condition on the face matching tests.