Menstrual cycle shifts in attentional bias for courtship language

Abstract

The current study investigated whether women show an attentional bias toward courtship language during the fertile phase of their menstrual cycle. Thirty heterosexual women (17 naturally cycling, 13 using hormonal contraceptives) completed a dichotic listening task on both a high and low fertility day of their menstrual cycle. Participants were asked to verbally repeat (shadow) an emotionally neutral target passage played in one ear while either a neutral or courtship distracter was played in the other ear. Courtship distracters were flirtatious in content but not overtly sexual. Shadowing errors were coded as a measure of attentional bias toward the distracter. Saliva samples were taken to determine whether levels of estradiol, progesterone and/or testosterone correlated with task performance. As predicted, naturally cycling women made more shadowing errors when listening to a courtship distracter during the fertile phase of their cycle than during the nonfertile phase. This effect was moderated by relationship status, such that fertile, mated women showed an attentional bias for courtship language but fertile single women did not. However, because of small sample sizes in the analysis, this relationship should be viewed as preliminary. Hormonal analysis revealed that higher levels of salivary estradiol predicted greater attentional bias toward courtship language in naturally cycling women. These results suggest that women's attention is drawn to verbal courtship signals when they are fertile, and that this shift is linked to increased estradiol release during the periovulatory phase.

Introduction

Unlike the majority of female mammals, women do not exhibit conspicuous morphological or behavioral changes around ovulation that signal their fertility to potential reproductive partners (Alexander and Noonan, 1979, Hrdy, 1981, Lovejoy, 1981, Pawlowski, 1999, Symons, 1979). However, numerous studies have found that women experience reliable changes in mating psychology and behavior across their menstrual cycle (for reviews, see Gangestad et al., 2005a, Regan, 1996, Tarín and Gómez-Piquer, 2002). For example, both heterosexual and homosexual women self-report an increase in initiated sexual encounters, autosexual behavior and number of orgasms experienced during their periovulatory phase, when the probability of conception rises (Adams et al., 1978, Burleson et al., 2002, Matteo and Rissman, 1984; but see Regan, 1996, for a discussion of the diversity of this literature, including null findings). More recent work has suggested that heterosexual women do not experience a general, periovulatory increase in sexual desire insomuch as increased interest in particular categories and characteristics of men. Gangestad et al., 2002, Gangestad et al., 2005b found that women in relationships reported greater sexual interest in men other than their primary partner near ovulation, while interest in their partner did not alter significantly. Women tested in their late-follicular phase express an increased attraction to masculine faces (Little et al., 2002, Penton-Voak et al., 1999, Penton-Voak and Perrett, 2000) and voice pitch (Feinberg et al., 2006), as well as the scent of highly symmetrical men (Thornhill & Gangestad, 1999). Gangestad and Cousins, 2001, Gangestad et al., 2005a, Gangestad et al., 2007 have argued that these psychological shifts are evidence of a mixed-mating strateg`y, in which women place greater emphasis on cues of high genetic fitness when seeking short-term sexual partners during the most fertile portion of their menstrual cycle.

This brief and incomplete review highlights several intriguing facets of female psychology. First, hormonal fluctuations across the menstrual cycle are capable of influencing mental processes related to mate selection. Second, these alterations in psychology are affected by social–environmental factors, such as relationship status. The process by which reproductive hormones and relevant mate information interact within specific neurocognitive mechanisms dedicated for courtship and mate assessment remains vague and undefined. Within this growing literature, several key issues need to be addressed:

• Studies measuring increases in sexual motivation have relied heavily on self-report methodologies, such as questionnaires and journal entries. This is problematic not only because self-report data are susceptible to memory distortion and falsification, but also because introspection is a poor strategy for gaining access to cognitive mechanisms that often function beyond individual self-consciousness.

• Recent studies in this field have focused predominantly on women's responses to male physical attributes, such as secondary sex characteristics and morphological symmetry. While physical characteristics clearly influence female mate choice, it can be argued that nonphysical, mental attributes are more important in human mate selection (Buss, 1989, Buss and Barnes, 1986). Humans are likely unique in their reliance upon language to communicate nonphysical aspects of fitness to potential mates during courtship. Experimental analyses of courtship language would shed insight into how humans display their mate value to one another, as well as express their own socio-sexual intentions.

• Finally, relatively few studies have attempted to determine which specific hormones regulate ovulatory shifts in women's sexual desire and mate preference. Women's preference for masculine cues, such as facial structure and voice pitch, has been negatively correlated with circulating levels of progesterone (Jones et al., 2005, Puts, 2006) and positively correlated with both testosterone (Welling et al., 2007) and estradiol (Roney & Simmons, 2008). Only the last two studies measured actual hormone concentrations in participants.

The current study was designed with these issues in mind. Specifically, we hypothesized that menstrual cycle fluctuations in reproductive hormones affect attentional bias for courtship-related language content. Attentional bias refers to a nonrandom shift in cognitive expenditure toward a particular category of environmental information. A functionalist perspective holds that cognitive and perceptual systems are designed to be particularly sensitive to input indicative of an adaptive challenge (Gibson, 1979, Maner et al., 2003, Ohman and Mineka, 2001). Several laboratories have successfully examined selective attention for other ecologically relevant categories, such as threat stimuli (Constantine et al., 2001, Hansen and Hansen, 1988, Ohman and Mineka, 2001). Attentional bias is theoretically modulated by underlying motivational systems, which can vary in valence across time and environmental circumstances and between individuals (Bauer and Cox, 1998, Bradley et al., 2003, Stetter et al., 1994). It is possible therefore that women experience functional adjustments to perceptual/attentional systems across their menstrual cycle, as reproductive hormones modify sexual motivational systems (Krug, 2000, Krug et al., 1994).

Humans express their romantic and sexual interest through flirtatious language, which often includes flattery (“You are beautiful”) and courtship intentions (“I would like to kiss you”). To assess attentional bias for specific language content, we adopted the dichotic listening and shadowing task (Barroso, 1983). A participant is played two sets of auditory stimuli simultaneously, one in each ear, and asked to attend to one while ignoring the other. The participant must repeat, or “shadow,” the target passage while ignoring the distracter content. If the distracter contains information that is salient to the participant (e.g., their name), it is harder to ignore and increases shadowing errors. Alexander and Sherwin (1993) used this task to examine distraction by sexual stimuli in women taking hormonal contraceptives. Participants made significantly more shadowing errors when the distracter passage was erotic in nature than when it was emotionally neutral. Furthermore, in women with subnormal testosterone levels, there was a positive correlation between plasma testosterone concentration and mistakes made due to the erotic distracter.

In the present study, both naturally cycling women (not taking hormonal contraceptives) and hormonal contraceptive users completed a dichotic listening and shadowing task at two points of their menstrual cycle. We hypothesized that naturally cycling women would be more distracted by flirtatious courtship language, and thus commit more shadowing errors, during the fertile portion of their menstrual cycle. We expected no such shift in women on hormonal contraceptives. In addition, all participants provided salivary samples during the experimental sessions, which were analyzed for estradiol, progesterone and testosterone content.