- Email: [email protected]
Do angry men get noticed?
Current Biology Vol 16 No 11 R402
A 30
20
10
0
0
0.04 0.08 0.12 0.16 0.20 0.24
0
0.04 0.08 0.12 0.16 0.20 0.24
0
0.04 0.08 0.12 0.16 0.20 0.24
B 30
20
10
0
C
30
20
10
0
Current Biology
Figure 2. Pairwise distance distribution between modern humans and Neandertals. The distributions of pairwise distances expressed in terms of number of substitutions per site are estimated from the best-fitting model, within the Homo sapiens sequences (A, n = 14535), between the Homo sapiens and the first six Neandertal sequences of Figure 1 except Scladina (B, n = 855), between the Homo sapiens and the Neandertal sequence from Scladina (C, n = 171). The y-axis shows the percentage of the pairwise counts. The arrows show the means of each distribution. The means of (B) and (C) (0.108 and 0.136, respectively) are significantly different (p < 0.001). The distributions in (A) and (B) overlap extensively: 95% (and 99%) of this distribution depicted in (A) overlap with 21% (and 69%) of the distribution in (B). The extent of this overlap is drastically reduced to 7% (and 21%) of the distribution in (C). Actually, 94% of the distribution in (A) does not overlap with (C). contribution to early modern humans. PLoS Biol. 2, E57. 5. Caramelli, D., Lalueza-Fox, C., Vernesi, C., Lari, M., Casoli, A., Mallegni, F., Chiarelli, B., et al. (2003). Evidence for a genetic discontinuity between Neandertals and 24,000-year-old anatomically modern Europeans. Proc. Natl. Acad. Sci. USA 100, 6593–6597. 6. Currat, M., and Excoffier, L. (2004). Modern humans did not admix with Neanderthals during their range expansion into Europe. PLoS Biol. 2, E421.
7. Gutierrez, G., Sanchez, D., and Marin, A. (2002). A reanalysis of the ancient mitochondrial DNA sequences recovered from Neandertal bones. Mol. Biol. Evol. 19, 1359–1366. 8. Bocherens, H., Billiou, D., and Mariotti, A. (1999). Palaeoenvironmental and palaeodietary implications of isotopic biochemistry of last interglacial Neanderthal and mammal bones in Scladina cave (Belgium). J. Arc. Sci. 26, 599–607. 9. Higham, T., Ramsey, C.B., Karavanic, I., Smith, F.H., and Trinkaus, E. (2006). Revised direct radiocarbon dating of the Vindija G1 Upper Paleolithic Neandertals. Proc. Natl. Acad. Sci. USA 103, 553–557. 10. Gilbert, M.T.P., Bandelt, H.J., Hofreiter, M., and Barnes, I. (2005). Assessing ancient DNA studies. Trends Ecol. Evol. 20, 541–544. 11. Hofreiter, M., Jaenicke, V., Serre, D., von Haeseler, A., and Pääbo, S. (2001). DNA sequences from multiple amplifications reveal artifacts induced by cytosine deamination in ancient DNA. Nucleic Acids Res. 29, 4793–4799. 12. Gilbert, M.T., Hansen, A.J., Willerslev, E., Rudbeck, L., Barnes, I., Lynnerup, N., and Cooper, A. (2003). Characterization of genetic miscoding lesions caused by postmortem damage. Am. J. Hum. Genet. 72, 48–61. 13. Orlando, L., Bonjean, D., Bocherens, H., Thénot, A., Argant, A., Otte, M., and Hänni, C. (2002). Ancient DNA and the population genetics of cave bears (Ursus spelaeus) through space and time. Mol. Biol. Evol. 19, 1920–1933. 14. Loreille, O., Orlando, L., Patou-Mathis M., Philippe, M., Taberlet, P., and Hänni, C. (2001). Ancient DNA analysis reveals divergence of the cave bear, Ursus spelaeus, and brown bear, Ursus arctos, lineages. Curr. Biol. 11, 200–203. 15. Orlando, L., Leonard, J., Thénot, A., Laudet, V., Guérin, C., and Hänni, C. (2003). Ancient DNA analysis reveals woolly rhino evolutionary relationships. Mol. Phylogenet. Evol. 28, 485–499. 16. Handt, O., Meyer, S., and von Haeseler, A. (1998). Compilation of human mtDNA control region sequences. Nucleic Acids Res. 26, 126–129. 17. Bower, M.A., Spencer, M., Matsumura, S., Nisbet, R.E., and Howe, C. (2005). How many clones need to be sequenced from a single forensic or ancient DNA sample in order to determine a reliable consensus sequences? Nucleic Acids Res. 33, 2549–2556. 1CNRS
UMR 5534, UCB Lyon1, Centre de Génétique Moléculaire et Cellulaire, Bât. G. Mendel, 16 Rue R. Dubois, 69622 Villeurbanne, France. 2Present Address: CNRS UMR 5161, INRA LA 1237, Laboratoire de Biologie Moléculaire de la Cellule, Ecole Normale Supérieure de Lyon, 46 Allée d’Italie, 69364 Lyon Cedex 07, France. 3INSERM U535, Génétique Epidémiologique et Structure des Populations Humaines, Bat. Leriche, Hôpital Paul Brousse, BP 1000, 94817 Villejuif Cedex, France. 4Direction de l’Archéologie, Ministère de la région Wallone, 1 rue des Brigades d’Irlande, 5100 Namur, Belgium. 5Scladina, Archéologie Andennaise ASBL, 339D rue Fond des Vaux, 5300 Sclayn, Belgium. 6Université de Liège, Service de Préhistoire, place du XX Août 7, Bat A1, 4000 Liège, Belgium. *E-mail: [email protected]
Do angry men get noticed? Mark A. Williams1,2 and Jason B. Mattingley1 In humans, the physical differences between the sexes are readily apparent, but possible cognitive and perceptual differences are less obvious. As social animals, humans have specialized mechanisms for recognizing facial expressions, but the extent to which these mechanisms are tuned to differences between male and female faces remains unclear. We measured the efficiency with which emotional expressions conveyed by male and female faces are detected by male and female observers. Angry male faces were detected significantly more rapidly by male than female observers. Moreover, detection of angry male faces by either male or female observers was scarcely affected by the addition of neutral distractor faces to the search display. Our findings are consistent with the notion of a perceptual system in both males and females that has evolved to rapidly detect aggression in males. In humans, evolution has resulted in marked differentiation between males and females [1,2], including differences in the structural and functional organization of the brain. These differences are reflected in patterns of cognitive and behavioural abilities [3]. For example, females tend to perform better than males at fine motor and perceptual discrimination tasks, whereas males are better at route- finding tasks [3]. Males are also physically larger and more aggressive than females, and so more likely to pose a physical threat [4]. Such physical differences between the sexes may in turn have shaped the cognitive processes involved in detecting threatening behaviour in others. Early detection of an angry facial expression, for example, might reduce the
Magazine R403
likelihood of an injurious or potentially fatal confrontation [5]. Similarly, detection of a fearful expression might warn of a potential threat in the immediate vicinity (reviewed in [5]). Although much emphasis has been placed on such cognitive and physical distinctions between the sexes, few studies have investigated differences in the efficiency with which males and females perceive facial expressions, despite the potential importance of affect perception for survival. Recent evidence suggests that females are better than males at recognizing non- threatening facial expressions such as happiness or sadness [3,6]. We compared the efficiency with which males and females detected threat-related or non- threat-related facial expressions in arrays of neutral distractor faces. From an evolutionary perspective, the potential for physical threat from a male is greater than that from a female [4]. A perceptual system that prioritizes detection of angry male faces, which directly signal potential threat, is therefore likely to be advantageous [5,7]. Our investigation focused on detection of threatening expressions in a large group of male (N = 78) and female (N = 78) participants. We presented an angry or fearful target face in an array of neutral distractor faces (Figure 1A). The comparison between angry and fearful expressions is of particular interest because, although both expressions signal potential threat, angry faces represent the source of a threat whereas fearful faces warn of possible danger elsewhere in the environment [8]. Face photographs of 12 individuals (6 male, 6 female), each displaying angry, fearful and neutral expressions, were selected from a standardized set of stimuli [9]. The faces were cropped using Photoshop 5.5 (Adobe Systems) to fit within an oval window subtending 1.6° x 2.2° of visual angle. Mean luminance and contrast were matched for all faces across all expressions. Either four or eight faces were
Figure 1. Example displays from a typical trial of the visual search task, and graphs of mean correct reaction time (±95% confidence intervals) for the two set sizes, plotted separately for angry and fearful face targets. Participants (78 male, 78 female) searched for an angry or fearful target face amongst neutral distractors, and indicated its presence or absence via a speeded keypress. (A) Example search displays containing an angry male target amongst three neutral distractor faces (left panel), and an angry female target amongst seven neutral distractor faces. There were equal numbers of male and female faces in each search display, and equal numbers of male and female targets across trials. Each trial began with a fixation point in the center of the screen for 1 second. This was followed by a search array, which remained visible until a response was made. Faces were positioned equidistant from the center of the display. Participants searched for the same target expression throughout a block of trials; set size was varied randomly within blocks. Each participant completed 160 trials (50% target present) per target expression. (B) Mean reaction time for male and female participants searching for a female target face. (C) Mean reaction time for male and female participants searching for a male target face.
positioned in a circle around fixation so that the tip of the nose was 5.2° from the center of the screen (Figure 1A). Male and female participants both detected angry faces significantly faster than fearful
faces, F(1, 152) = 36.72, p < 0.001 (Cohen’s d = 0.52). Crucially, however, this advantage for angry expressions depended on the sex of both the target face and the participant. The time required to detect female
Current Biology Vol 16 No 11 R404
faces, whether angry or fearful, increased uniformly with an increase in set size from four to eight items (p < 0.01; Figure 1B). In contrast, although detection of male faces also increased with set size for fearful expressions, there was no set size effect for angry male faces (p > 0.05; statistical power = 0.89; Figure 1C). The absence of a set size effect for angry male faces was confirmed by a highly significant three-way interaction between Target Sex, Target Expression and Set Size, F(1, 152) = 10.670, p < 0.001. Search times also varied according to the sex of the participant: males detected angry faces faster than females, despite equivalent search times between the sexes for fearful faces (Target Expression x Participants’ Sex, F(1, 154) = 5.790, p < 0.01; Target Expression x Participants’ Sex x Set Size, F(1, 152) = 27.971, p < 0.001; compare Figures 1B,C). A post-hoc t-test verified that males were significantly faster than females in detecting an angry male face (p < 0.01; Cohen’s d = 0.56); however, there was no difference in the speed with which male and female participants detected fearful faces (p> 0.05; statistical power = 0.82). This finding suggests that any threat detection system for angry male faces is more pronounced in males than in females. To rule out the effects of low- level featural differences between expression types, we also examined search performance for identical displays that were inverted to attenuate facial expression cues [10]. There were no significant differences across the factors of Target Sex, Participant Sex or Target Expression, demonstrating that the effects for upright faces cannot be attributed to featural differences between the sexes or expression types. Further, there were no significant priming effects due to the repetition of a particular face (F(1, 152) = .96, p > .05; see Figure S1 in the Supplemental data available on- line).
We also examined participants’ search performance for a neutral target face amongst either fearful of angry facial expressions. There was a significant main effect of Distractor Expression, (F(1, 154) = 45.86, p < 0.05) and a significant main effect of Set Size (F(1, 154) = 202.22, p < 0.05). Crucially, however, none of the other main effects or interactions were significant, in contrast to the results of our main experiment (see Figure S2 in the Supplemental data). Examination of search performance for the remaining primary facial expressions — happiness, sadness, surprise and disgust — found no reliable differences as a function of sex of the target. Interestingly, female participants detected targets faster than male participants for every expression except those indicating possible threat (see Supplemental data). Our results are consistent with the idea that evolution has refined and differentiated a threat-detection system in both males and females, for several reasons. First, angry male faces are prioritized by individuals of both sexes, as evidenced by the absence of any increase in search time with an increase in the number of neutral-face distractors. Second, males detected angry faces faster than females suggesting that the male visual system is more attuned to angry faces than the female system. Third, females detected the non- threat related emotions of happiness, sadness, surprise and disgust faster than males. This pattern of results suggests that, although males are biased towards detecting- threatening faces, and females are more attuned to socially relevant expressions, both sexes prioritize the detection of angry male faces; in short, angry men get noticed. The advantage for detecting angry male faces is consistent with the notion that human perceptual processes have been shaped by evolutionary pressures arising from the social environment [1,5].
Acknowledgements We would like to thank Anina Rich for her suggestions on an earlier draft of this manuscript. M.A.W. is a C.J. Martin Fellow sponsored by the Australian National Health and Medical Research Council. Supplemental data Supplemental data are available at http://www.current-biology.com/cgi/content/full/16/11/R402/DC1/ References 1. Anderson, M. (1994). Sexual Selection (New Jersey: Princeton University Press). 2. Darwin, C. (1871, 1896). The Descent of Man and Selection in Relation to Sex (New York: Appleton and Company). 3. Geary, D.C. (1998). Male, Female: The Evolution of Human Sex Differences (Washington, DC: American Psychological Association.). 4. Nelson, R.J., and Chiavegatto, S. (2001). Molecular basis of aggression. Trends Neurosci. 24, 713–719. 5. Öhman, A. (2002). Automaticity and the amygdala: nonconscious responses to emotional faces. Curr. Direct. Psychol. Sci. 11, 62–66. 6. Rahman, Q., Wilson, G.D., and Abrahams, S. (2004). Sex, sexual orientation, and isentification of positive and negative facial affect. Brain Cogn. 54, 179–185. 7. Fox, E. (2005). The role of visual processes in modulating social interactions. Visual Cogn. 12, 1–11. 8. Whalen, P.J. (1998). Fear, vigilance and ambiguity: initial neuroimaging studies of the human amygdala. Curr. Direct. Psychol. Sci. 7, 177–188. 9. Tottenham, N., Borscheid, A., Ellertsen, K., Marcus, D.J., and Nelson, C.A. (2002). Categorization of facial expressions in children and adults: establishing a larger stimulus set. In Cognitive Neuroscience Society Annual Meeting, (Poster, ed.): San Francisco, USA. 10. Williams, M.A., Moss, S.A., Bradshaw, J.L., and Mattingley, J.B. (2005). Look at me, I’m smiling: searching for threatening and non-threatening facial expressions. Visual Cogn. 12, 29–50. 1McGovern
Institute for Brain Research, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139, USA. 2Cognitive Neuroscience Laboratory, School of Behavioural Science, University of Melbourne, Parkville, Victoria 3010, Australia. E-mail: [email protected]
The editors of Current Biology welcome correspondence on any article in the journal, but reserve the right to reduce the length of any letter to be published. All Correspondence containing data or scientific argument will be refereed. Queries about articles for consideration in this format should be sent by e-mail to cbiol@ current-biology.com
A 30
20
10
0
0
0.04 0.08 0.12 0.16 0.20 0.24
0
0.04 0.08 0.12 0.16 0.20 0.24
0
0.04 0.08 0.12 0.16 0.20 0.24
B 30
20
10
0
C
30
20
10
0
Current Biology
Figure 2. Pairwise distance distribution between modern humans and Neandertals. The distributions of pairwise distances expressed in terms of number of substitutions per site are estimated from the best-fitting model, within the Homo sapiens sequences (A, n = 14535), between the Homo sapiens and the first six Neandertal sequences of Figure 1 except Scladina (B, n = 855), between the Homo sapiens and the Neandertal sequence from Scladina (C, n = 171). The y-axis shows the percentage of the pairwise counts. The arrows show the means of each distribution. The means of (B) and (C) (0.108 and 0.136, respectively) are significantly different (p < 0.001). The distributions in (A) and (B) overlap extensively: 95% (and 99%) of this distribution depicted in (A) overlap with 21% (and 69%) of the distribution in (B). The extent of this overlap is drastically reduced to 7% (and 21%) of the distribution in (C). Actually, 94% of the distribution in (A) does not overlap with (C). contribution to early modern humans. PLoS Biol. 2, E57. 5. Caramelli, D., Lalueza-Fox, C., Vernesi, C., Lari, M., Casoli, A., Mallegni, F., Chiarelli, B., et al. (2003). Evidence for a genetic discontinuity between Neandertals and 24,000-year-old anatomically modern Europeans. Proc. Natl. Acad. Sci. USA 100, 6593–6597. 6. Currat, M., and Excoffier, L. (2004). Modern humans did not admix with Neanderthals during their range expansion into Europe. PLoS Biol. 2, E421.
7. Gutierrez, G., Sanchez, D., and Marin, A. (2002). A reanalysis of the ancient mitochondrial DNA sequences recovered from Neandertal bones. Mol. Biol. Evol. 19, 1359–1366. 8. Bocherens, H., Billiou, D., and Mariotti, A. (1999). Palaeoenvironmental and palaeodietary implications of isotopic biochemistry of last interglacial Neanderthal and mammal bones in Scladina cave (Belgium). J. Arc. Sci. 26, 599–607. 9. Higham, T., Ramsey, C.B., Karavanic, I., Smith, F.H., and Trinkaus, E. (2006). Revised direct radiocarbon dating of the Vindija G1 Upper Paleolithic Neandertals. Proc. Natl. Acad. Sci. USA 103, 553–557. 10. Gilbert, M.T.P., Bandelt, H.J., Hofreiter, M., and Barnes, I. (2005). Assessing ancient DNA studies. Trends Ecol. Evol. 20, 541–544. 11. Hofreiter, M., Jaenicke, V., Serre, D., von Haeseler, A., and Pääbo, S. (2001). DNA sequences from multiple amplifications reveal artifacts induced by cytosine deamination in ancient DNA. Nucleic Acids Res. 29, 4793–4799. 12. Gilbert, M.T., Hansen, A.J., Willerslev, E., Rudbeck, L., Barnes, I., Lynnerup, N., and Cooper, A. (2003). Characterization of genetic miscoding lesions caused by postmortem damage. Am. J. Hum. Genet. 72, 48–61. 13. Orlando, L., Bonjean, D., Bocherens, H., Thénot, A., Argant, A., Otte, M., and Hänni, C. (2002). Ancient DNA and the population genetics of cave bears (Ursus spelaeus) through space and time. Mol. Biol. Evol. 19, 1920–1933. 14. Loreille, O., Orlando, L., Patou-Mathis M., Philippe, M., Taberlet, P., and Hänni, C. (2001). Ancient DNA analysis reveals divergence of the cave bear, Ursus spelaeus, and brown bear, Ursus arctos, lineages. Curr. Biol. 11, 200–203. 15. Orlando, L., Leonard, J., Thénot, A., Laudet, V., Guérin, C., and Hänni, C. (2003). Ancient DNA analysis reveals woolly rhino evolutionary relationships. Mol. Phylogenet. Evol. 28, 485–499. 16. Handt, O., Meyer, S., and von Haeseler, A. (1998). Compilation of human mtDNA control region sequences. Nucleic Acids Res. 26, 126–129. 17. Bower, M.A., Spencer, M., Matsumura, S., Nisbet, R.E., and Howe, C. (2005). How many clones need to be sequenced from a single forensic or ancient DNA sample in order to determine a reliable consensus sequences? Nucleic Acids Res. 33, 2549–2556. 1CNRS
UMR 5534, UCB Lyon1, Centre de Génétique Moléculaire et Cellulaire, Bât. G. Mendel, 16 Rue R. Dubois, 69622 Villeurbanne, France. 2Present Address: CNRS UMR 5161, INRA LA 1237, Laboratoire de Biologie Moléculaire de la Cellule, Ecole Normale Supérieure de Lyon, 46 Allée d’Italie, 69364 Lyon Cedex 07, France. 3INSERM U535, Génétique Epidémiologique et Structure des Populations Humaines, Bat. Leriche, Hôpital Paul Brousse, BP 1000, 94817 Villejuif Cedex, France. 4Direction de l’Archéologie, Ministère de la région Wallone, 1 rue des Brigades d’Irlande, 5100 Namur, Belgium. 5Scladina, Archéologie Andennaise ASBL, 339D rue Fond des Vaux, 5300 Sclayn, Belgium. 6Université de Liège, Service de Préhistoire, place du XX Août 7, Bat A1, 4000 Liège, Belgium. *E-mail: [email protected]
Do angry men get noticed? Mark A. Williams1,2 and Jason B. Mattingley1 In humans, the physical differences between the sexes are readily apparent, but possible cognitive and perceptual differences are less obvious. As social animals, humans have specialized mechanisms for recognizing facial expressions, but the extent to which these mechanisms are tuned to differences between male and female faces remains unclear. We measured the efficiency with which emotional expressions conveyed by male and female faces are detected by male and female observers. Angry male faces were detected significantly more rapidly by male than female observers. Moreover, detection of angry male faces by either male or female observers was scarcely affected by the addition of neutral distractor faces to the search display. Our findings are consistent with the notion of a perceptual system in both males and females that has evolved to rapidly detect aggression in males. In humans, evolution has resulted in marked differentiation between males and females [1,2], including differences in the structural and functional organization of the brain. These differences are reflected in patterns of cognitive and behavioural abilities [3]. For example, females tend to perform better than males at fine motor and perceptual discrimination tasks, whereas males are better at route- finding tasks [3]. Males are also physically larger and more aggressive than females, and so more likely to pose a physical threat [4]. Such physical differences between the sexes may in turn have shaped the cognitive processes involved in detecting threatening behaviour in others. Early detection of an angry facial expression, for example, might reduce the
Magazine R403
likelihood of an injurious or potentially fatal confrontation [5]. Similarly, detection of a fearful expression might warn of a potential threat in the immediate vicinity (reviewed in [5]). Although much emphasis has been placed on such cognitive and physical distinctions between the sexes, few studies have investigated differences in the efficiency with which males and females perceive facial expressions, despite the potential importance of affect perception for survival. Recent evidence suggests that females are better than males at recognizing non- threatening facial expressions such as happiness or sadness [3,6]. We compared the efficiency with which males and females detected threat-related or non- threat-related facial expressions in arrays of neutral distractor faces. From an evolutionary perspective, the potential for physical threat from a male is greater than that from a female [4]. A perceptual system that prioritizes detection of angry male faces, which directly signal potential threat, is therefore likely to be advantageous [5,7]. Our investigation focused on detection of threatening expressions in a large group of male (N = 78) and female (N = 78) participants. We presented an angry or fearful target face in an array of neutral distractor faces (Figure 1A). The comparison between angry and fearful expressions is of particular interest because, although both expressions signal potential threat, angry faces represent the source of a threat whereas fearful faces warn of possible danger elsewhere in the environment [8]. Face photographs of 12 individuals (6 male, 6 female), each displaying angry, fearful and neutral expressions, were selected from a standardized set of stimuli [9]. The faces were cropped using Photoshop 5.5 (Adobe Systems) to fit within an oval window subtending 1.6° x 2.2° of visual angle. Mean luminance and contrast were matched for all faces across all expressions. Either four or eight faces were
Figure 1. Example displays from a typical trial of the visual search task, and graphs of mean correct reaction time (±95% confidence intervals) for the two set sizes, plotted separately for angry and fearful face targets. Participants (78 male, 78 female) searched for an angry or fearful target face amongst neutral distractors, and indicated its presence or absence via a speeded keypress. (A) Example search displays containing an angry male target amongst three neutral distractor faces (left panel), and an angry female target amongst seven neutral distractor faces. There were equal numbers of male and female faces in each search display, and equal numbers of male and female targets across trials. Each trial began with a fixation point in the center of the screen for 1 second. This was followed by a search array, which remained visible until a response was made. Faces were positioned equidistant from the center of the display. Participants searched for the same target expression throughout a block of trials; set size was varied randomly within blocks. Each participant completed 160 trials (50% target present) per target expression. (B) Mean reaction time for male and female participants searching for a female target face. (C) Mean reaction time for male and female participants searching for a male target face.
positioned in a circle around fixation so that the tip of the nose was 5.2° from the center of the screen (Figure 1A). Male and female participants both detected angry faces significantly faster than fearful
faces, F(1, 152) = 36.72, p < 0.001 (Cohen’s d = 0.52). Crucially, however, this advantage for angry expressions depended on the sex of both the target face and the participant. The time required to detect female
Current Biology Vol 16 No 11 R404
faces, whether angry or fearful, increased uniformly with an increase in set size from four to eight items (p < 0.01; Figure 1B). In contrast, although detection of male faces also increased with set size for fearful expressions, there was no set size effect for angry male faces (p > 0.05; statistical power = 0.89; Figure 1C). The absence of a set size effect for angry male faces was confirmed by a highly significant three-way interaction between Target Sex, Target Expression and Set Size, F(1, 152) = 10.670, p < 0.001. Search times also varied according to the sex of the participant: males detected angry faces faster than females, despite equivalent search times between the sexes for fearful faces (Target Expression x Participants’ Sex, F(1, 154) = 5.790, p < 0.01; Target Expression x Participants’ Sex x Set Size, F(1, 152) = 27.971, p < 0.001; compare Figures 1B,C). A post-hoc t-test verified that males were significantly faster than females in detecting an angry male face (p < 0.01; Cohen’s d = 0.56); however, there was no difference in the speed with which male and female participants detected fearful faces (p> 0.05; statistical power = 0.82). This finding suggests that any threat detection system for angry male faces is more pronounced in males than in females. To rule out the effects of low- level featural differences between expression types, we also examined search performance for identical displays that were inverted to attenuate facial expression cues [10]. There were no significant differences across the factors of Target Sex, Participant Sex or Target Expression, demonstrating that the effects for upright faces cannot be attributed to featural differences between the sexes or expression types. Further, there were no significant priming effects due to the repetition of a particular face (F(1, 152) = .96, p > .05; see Figure S1 in the Supplemental data available on- line).
We also examined participants’ search performance for a neutral target face amongst either fearful of angry facial expressions. There was a significant main effect of Distractor Expression, (F(1, 154) = 45.86, p < 0.05) and a significant main effect of Set Size (F(1, 154) = 202.22, p < 0.05). Crucially, however, none of the other main effects or interactions were significant, in contrast to the results of our main experiment (see Figure S2 in the Supplemental data). Examination of search performance for the remaining primary facial expressions — happiness, sadness, surprise and disgust — found no reliable differences as a function of sex of the target. Interestingly, female participants detected targets faster than male participants for every expression except those indicating possible threat (see Supplemental data). Our results are consistent with the idea that evolution has refined and differentiated a threat-detection system in both males and females, for several reasons. First, angry male faces are prioritized by individuals of both sexes, as evidenced by the absence of any increase in search time with an increase in the number of neutral-face distractors. Second, males detected angry faces faster than females suggesting that the male visual system is more attuned to angry faces than the female system. Third, females detected the non- threat related emotions of happiness, sadness, surprise and disgust faster than males. This pattern of results suggests that, although males are biased towards detecting- threatening faces, and females are more attuned to socially relevant expressions, both sexes prioritize the detection of angry male faces; in short, angry men get noticed. The advantage for detecting angry male faces is consistent with the notion that human perceptual processes have been shaped by evolutionary pressures arising from the social environment [1,5].
Acknowledgements We would like to thank Anina Rich for her suggestions on an earlier draft of this manuscript. M.A.W. is a C.J. Martin Fellow sponsored by the Australian National Health and Medical Research Council. Supplemental data Supplemental data are available at http://www.current-biology.com/cgi/content/full/16/11/R402/DC1/ References 1. Anderson, M. (1994). Sexual Selection (New Jersey: Princeton University Press). 2. Darwin, C. (1871, 1896). The Descent of Man and Selection in Relation to Sex (New York: Appleton and Company). 3. Geary, D.C. (1998). Male, Female: The Evolution of Human Sex Differences (Washington, DC: American Psychological Association.). 4. Nelson, R.J., and Chiavegatto, S. (2001). Molecular basis of aggression. Trends Neurosci. 24, 713–719. 5. Öhman, A. (2002). Automaticity and the amygdala: nonconscious responses to emotional faces. Curr. Direct. Psychol. Sci. 11, 62–66. 6. Rahman, Q., Wilson, G.D., and Abrahams, S. (2004). Sex, sexual orientation, and isentification of positive and negative facial affect. Brain Cogn. 54, 179–185. 7. Fox, E. (2005). The role of visual processes in modulating social interactions. Visual Cogn. 12, 1–11. 8. Whalen, P.J. (1998). Fear, vigilance and ambiguity: initial neuroimaging studies of the human amygdala. Curr. Direct. Psychol. Sci. 7, 177–188. 9. Tottenham, N., Borscheid, A., Ellertsen, K., Marcus, D.J., and Nelson, C.A. (2002). Categorization of facial expressions in children and adults: establishing a larger stimulus set. In Cognitive Neuroscience Society Annual Meeting, (Poster, ed.): San Francisco, USA. 10. Williams, M.A., Moss, S.A., Bradshaw, J.L., and Mattingley, J.B. (2005). Look at me, I’m smiling: searching for threatening and non-threatening facial expressions. Visual Cogn. 12, 29–50. 1McGovern
Institute for Brain Research, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139, USA. 2Cognitive Neuroscience Laboratory, School of Behavioural Science, University of Melbourne, Parkville, Victoria 3010, Australia. E-mail: [email protected]
The editors of Current Biology welcome correspondence on any article in the journal, but reserve the right to reduce the length of any letter to be published. All Correspondence containing data or scientific argument will be refereed. Queries about articles for consideration in this format should be sent by e-mail to cbiol@ current-biology.com