Report

Human Tears Contain a Chemosignal

See allHide authors and affiliations

Science  14 Jan 2011:
Vol. 331, Issue 6014, pp. 226-230
DOI: 10.1126/science.1198331

You are currently viewing the figures only.

View Full Text

Log in to view the full text

Log in through your institution

Log in through your institution

  1. Fig. 1

    Emotional tears are odorless. (A) To obtain tears, donor women watched sad films in isolation, using a mirror to capture tears into a vial. A typical donation contained ~1 ml. (B) For continuous exposure, tears deposited onto a pad allowed ongoing nasal airflow exposure but not transdermal diffusion. (A) and (B) are illustrations. (C) Accuracy of discrimination of tears from saline. (D to F) Scatterplots of (D) intensity, (E) pleasantness, and (F) familiarity estimates of tears and saline in all experiments. Within-participants comparisons [(D) to (F) here and in all other figures] are presented in scatterplots along a unit slope line (x = y), where each point reflects a participant. If data accumulate under the line, then values were greater for tears; if data accumulate above the line, then values were greater for saline. If data accumulate on the line, then there was no difference between tears and saline. (Inset) Bars reflect the number of participants on each side of the unit slope line (left ordinate), and the horizontal dashed line reflects the mean values and standard error (right ordinate).

  2. Fig. 2

    Sniffing tears reduces attributed sexual attraction. (A) and (B) Typical VAS questions from the face-rating experiment. (C) Attributed sexual attraction by 24 men. Data accumulated above the line, indicating reduced attributed sexual attraction after sniffing tears. (Inset) Bars reflect the number of participants on each side of the unit slope line (left ordinate), and the horizontal dashed line reflects the mean values and standard error (right ordinate).

  3. Fig. 3

    Sniffing tears reduces arousal. (A) Shift in self-rated sexual arousal from baseline to ~5 min after sniffing tears, indicating a drop in arousal. (B) Shift in self-rated sexual arousal from baseline to ~5 min after sniffing saline, indicating no change. (C and E) Timeline of psychophysiological data. Data were aligned in time to the first sniff (dashed vertical line). The abscissa lists the time in seconds and the experimental phases. BL, baseline; Q/S, VAS mood questionnaire, followed by a saliva sample; EE, experimenter enters room. Experimental phases are approximately (~) accurate in time, because participants differed in questionnaire latency, resulting in a shift of a few seconds. Error bars reflect between-participants variance, here displayed only to give a better sense of the data. The statistically relevant variance is the within-participants variance displayed in scatterplots (D) and (F). (C) Timeline of ongoing GSR amplitude, indicating greater GSR response for tears during sniffing only. (D) GSR change from sniffing to the end of a sad movie, indicating greater change for tears. (E) Timeline of the CPI, indicating reduced arousal after sniffing tears. (F) CPI at the final stage of the study, indicating significantly lower arousal after sniffing tears versus saline. (G) Shift in salivary levels of testosterone from baseline to last saliva collection after sniffing tears, indicating a drop from baseline. (H) Shift in salivary levels of testosterone from baseline to last saliva collection after sniffing saline, indicating no change. (Insets) Bars reflect the number of participants on each side of the unit slope line (left ordinate), and the horizontal dashed line reflects the mean values and standard error (right ordinate).

  4. Fig. 4

    Sniffing tears reduces brain activity in substrates of sexual arousal. (A) Activity induced by an erotic film generated a region of interest (ROI) in the hypothalamus. (B) Average activity time-course from 16 men within the previously identified hypothalamic ROI. This activity was induced by a sad film clip after sniffing either tears (blue) or saline (red). (C) Area under the curve of activation from (B), providing a measure of variance across participants. (D) Activity induced by an erotic film generated an ROI in the left fusiform gyrus. (E) Average activity time-course from 16 men within the previously identified fusiform gyrus ROI. This activity was induced by a sad film clip after sniffing either tears (blue) or saline (red). (F) Area under the curve of activation from (E), providing a measure of variance across participants. (Insets) Bars reflect the number of participants on each side of the unit slope line (left ordinate), and the horizontal dashed line reflects the mean values and standard error (right ordinate).