Relation between comparative risk,absolute risk,and worry: The role of handedness strength

Abstract

Research is mixed regarding whether comparative or absolute risk perceptions are most associated with important outcomes (e.g. worry). The current research examined the role of individual differences in organizational brain structure and connectivity—assessed via handedness strength—in moderating these effects. Strong handers and mixed handers made comparative and absolute risk estimates for contracting a sexually transmitted disease or sexually transmitted infection and also indicated their worry. Additionally, participants indicated whether they received testing for sexually transmitted diseases or sexually transmitted infections. For strong handers, absolute risk perceptions were generally more associated with worry and testing behaviors, whereas for mixed handers, comparative risk perceptions were more associated.

Keywords

absolute versus comparative risk handedness risk perception social comparison worry

Risk perceptions are important variables in health behavior models (e.g. health belief model; Janz and Becker, 1984). Many investigations have focused on the connection between risk and worry, since worry is often conceptualized as a precursor to adaptive cognition and behavior (Davey, 1993; Leventhal et al., 2003; Miller et al., 1996) and can be impacted by risk (Chapman and Coups, 2006; Claassen et al., 2010; French et al., 2004, 2006; Klein, 1997; McCaul and Mullens, 2003; Mason et al., 2008; Rose, in press). Complicating the development of these models, there are numerous ways to conceptualize risk.

The vast majority of research involves absolute risk or verbal (“very unlikely”) and numerical (“20% likely”) likelihood estimates for experiencing an outcome. However, contemporary research has examined social comparative risk or whether a person’s likelihood is less, the same, or more than a specified reference group. When pitted against one another, there is mixed evidence regarding whether absolute or comparative risk is more associated with worry and other variables (e.g. Blalock et al., 1990; Dillard et al., 2012; French et al., 2004, 2006; Harris et al., 2002; Harris and Smith, 2005; Klein, 1997, 2002; Lipkus et al., 2000; Mason et al., 2008; Rose, 2010). Although differences in methodologies, samples, and risk domains have been noted as potential explanations, no definitive answers currently exist. One possibility is that systematic individual differences across or within studies can explain why some people’s worry is more highly correlated with absolute versus comparative risk (Zajac et al., 2006). The main goal of the current research was to examine the moderating role of individual differences in organizational brain structure and connectivity—assessed via handedness strength.

Handedness, interhemispheric interaction, risk, and worry

Handedness has a rich history in psychology as an important individual difference. Although previous research has focused on direction (left or right), a more recent movement suggests that handedness strength may be more important (e.g. Christman et al., 2007; Jasper and Christman, 2005; Jasper et al., 2008; Rose et al., 2012). In this research, mixed handers (MH), who use their nondominant hand for at least some activities, are distinguished from strong handers (SH), who use one dominant hand for all activities. Conceptually speaking, handedness strength has been interpreted in terms of increased interaction between the left and right cerebral hemispheres and/or as greater access to right hemisphere–based processing. This interpretation is based on morphological studies illustrating that MH tend to have larger corpora callosa—the bundle of nerves connecting the left and right hemispheres (Clarke and Zaidel, 1994; Witelson and Goldsmith, 1991). Moreover, converging neurological and behavioral evidence across various contexts (e.g. episodic memory) supports greater interhemispheric interaction between left and right hemisphere–based processes among MH (Cabeza and Nyberg, 2000; Propper et al., 2005).

We suggest that MH (vs SH) should show more sensitivity to social comparative risk for two interrelated reasons. First, MH have more access to socially relevant information than do SH. For instance, representations of others (Feinberg and Keenan, 2005) and perspective taking (Perry et al., 2001; Rankin et al., 2006) are localized within the right hemisphere (e.g. right temporoparietal junction; Ruby and Decety, 2004). Moreover, damage to the right hemisphere is associated with impairment on perspective taking tasks relative to healthy controls and those with left hemisphere damage (Happe et al., 1999). Due to their increased access to right hemisphere–based processes, MH should have better access to social comparative information and hence show a greater association between comparative risk and worry.

Second, MH are more likely to use social comparative information. Rose et al. (2012) showed that MH are less egocentric when judging their comparative standing relative to others. For instance, when asked to evaluate themselves relative to others, SH showed the typical egocentric pattern wherein they believed that they were better-than-average (BTA) for easy tasks but worse-than-average (WTA) for hard tasks. That is, SH failed to consider that others are similarly skilled and unskilled at such tasks. However, MH showed less egocentrism and a lower magnitude of BTA and WTA biases—demonstrating greater utilization of social comparative information. This suggests that MH may be more likely to weight comparative information when contextualizing their worry.

Current research

Our primary goal was to examine whether handedness strength moderates the association between risk and worry. College students made absolute risk estimates, comparative risk estimates, and worry estimates about contracting a sexually transmitted disease or sexually transmitted infection (STD/STI). We chose this risk context because it was deemed more relevant to this population than traditional health outcomes (e.g. heart attack) and because there are social implications. We hypothesized that MH—who have a greater degree of interhemispheric connectivity and more access to the right hemisphere—would show a greater association between comparative risk and worry than would SH, where being higher in comparative risk would be associated with greater worry. Moreover, we also assessed participants’ STD/STI testing behaviors for both the initial survey and again later in the semester. Thus, a secondary goal was to examine whether handedness strength might have a similar moderating impact on the relationship between risk and subsequent testing behaviors.

Method

Participants and design

Eighty-five students (45 females) from a large Midwestern University in the United States participated for a course requirement.

Measures and procedure

Participants completed the study online as part of a larger study involving individual differences, personality, attitudes, and risk. First, for comparative risk estimates, participants were asked: “Compared to the average student at your university of the same age/sex, how likely is it that you will get a sexually-transmitted disease or infection?” (1 = much less likely than the average student; 11 = much more likely than the average student). Second, for verbal and numeric absolute risk estimates, participants were asked: “What is the chance that you will get a sexually-transmitted disease or infection?” (1 = not at all likely; 11 = very likely) and “What is the likelihood that you will get a sexually-transmitted disease or infection” (answered on 11-point scales with 0% and 100% as anchors). Third, participants answered the two questions related to how worried and concerned they were about contracting a STD/STI (1 = not at all; 11 = very; α = .89). Moreover, participants also indicated whether or not they had been tested for STDs/STIs (0 = no; 1 = yes). More critically, later in the semester during an unrelated study, participants were again asked whether they had been tested for STDs/STIs.

Using the Edinburgh Handedness Inventory (EHI; Oldfield, 1971), participants indicated their hand use/preference for 10 common activities (e.g. throwing, using a spoon; α = .89) using a 5-point scale (“Always Left,” “Usually Left,” “No Preference,” “Usually Right,” or “Always Right”). Scores can range from −100 (strongly left handed) to +100 (strongly right handed). Following the practice of previous research (Christman et al., 2007; Jasper and Christman, 2005; Jasper et al., 2008), the median absolute score in the sample (85) was used as a cutoff score to define SH versus MH. Participants scoring at or above 85 (absolute value) were categorized as strong handed (n = 41), and those scoring below 85 were categorized as mixed handed (n = 44).

Results

Association between risk perceptions and worry

To examine whether the association between risk estimates and worry differed as a function of handedness, we used a moderated regression approach. Specifically, after centering, we regressed worry onto absolute verbal risk estimates, absolute numeric risk estimates, social comparative risk estimates, and handedness (1 = mixed handed; 2 = strong handed; see Table 1 for means and standard deviations (SDs)). The first step included all main effects, and the second step included the critical two-way interactions (see Table 2).

Table 1.

Risk perceptions, worry, and testing status as a function of handedness strength.

Measure	MH		SH		Inferential
	M	SD	M	SD	t	p
Comparative	2.51	2.16	2.61	2.17	−.22	.83
Absolute verbal	2.10	1.48	2.43	2.12	−.84	.41
Absolute numeric	11.95	15.53	14.09	20.15	−.54	.59
Worry	2.57	1.90	2.84	2.64	−.53	.60
Testing status T1	.37	.49	.31	.47	.54	.60
Testing status T2	.41	.50	.36	.48	.49	.63

MH: mixed handers; SH: strong handers; SD: standard deviation; STD: sexually transmitted disease; STI: sexually transmitted infection.

Note: All worry and risk estimates were made on 11-point scales, with the following anchors for comparative estimates (1 = much less likely than the average student; 11 = much more likely than the average student), absolute verbal estimates (1 = not at all likely; 11 = very likely), absolute numeric estimates (0-100% likely), and worry (1 = not at all; 11 = very). The testing status variable asked whether or not students have been tested for STDs/STIs (0 = no; 1 = yes). Values in the “Inferential statistics” column provide t and p values from independent sample t-tests comparing MH and SH.

Table 2.

Hierarchical moderated regression analyses for the relationship between risk perceptions and worry.

Variable	β	R ²	R²Δ
Step 1: main effects
Absolute verbal	.58**	.20**	.20**
Absolute numeric	−.25
Social comparative	.09
Handedness	.08
Step 2: two-way interactions
Absolute verbal × handedness	.46*	.37**	.17**
Absolute numeric × handedness	−.22
Social comparative × handedness	−.24*
Absolute verbal × absolute numeric	−.55**
Absolute verbal × social comparative	.05
Absolute numeric × social comparative	−.05

p < .01; *p < .05.

First, the main effect model accounted for 20 percent of the variance in worry, F(4, 80) = 4.94, p < .01. The only significant predictor was absolute verbal risk estimates, β = .57, t(83) = 2.87, p < .01, indicating that greater risk was associated with more worry (all other βs < |.25|, ts < 1.4, ps > .15). Second, the two-way interaction model accounted for unique variance in worry, R²Δ = .17, F(6, 74) = 3.26, p < .01. We had a priori interest in examining the three interactions involving handedness strength. First, both the absolute verbal risk × handedness strength and the social comparative risk × handedness strength interactions were significant (βs > |.24|, ts > |2.02|, ps < .05). The absolute numeric risk × handedness strength interaction was not significant (β = −.22, t = −1.17, p > .20; all other two-way interactions were not significant, βs < |.05|, ts < .20, ps > .10, except the absolute verbal × absolute numeric interaction, β = −.55, t = −2.83, p < .01). To interpret the two most theoretically meaningful (and significant) interactions, worry estimates were graphed as a function of handedness strength at 1 SD above and below the mean of risk estimates (see Figure 1). First, worry appeared more impacted by social comparative risk estimates in MH than in SH, where higher comparative risk estimates were associated with more worry. Second, worry appeared more impacted by absolute verbal risk estimates in SH than in MH, where higher absolute risk estimates were associated with more worry.¹

Figure 1.

Worry as a function of handedness strength, absolute verbal risk, and comparative risk.

Association between risk perceptions and STD/STI testing

To examine the relationship between risk estimates and STD/STI testing, we again used a moderated regression approach. To control for initial testing status, the first step regressed testing status later in the semester onto testing status at the beginning of the semester (0 = no; 1 = yes). In the second step, after centering, we regressed testing status onto absolute verbal risk estimates, absolute numeric risk estimates, social comparative risk estimates, and handedness (1 = mixed handed; 2 = strong handed). The third step included the two-way interactions (see Table 3).

Table 3.

Hierarchical moderated regression analyses for the relationship between risk perceptions and STD/STI testing status.

Variable	β	R ²	R²Δ
Step 1: prior testing status
Prior testing status	.84**	.71**	.71**
Step 2: main effects
Absolute verbal	−.02	.72**	.01
Absolute numeric	−.06
Social comparative	.11
Handedness	−.002
Step 3: two-way interactions
Absolute verbal × handedness	.27†	.74**	.02
Absolute numeric × handedness	−.07
Social comparative × handedness	−.16*
Absolute verbal × absolute numeric	−.07
Absolute verbal × social comparative	.07
Absolute numeric × social comparative	−.07

STD: sexually transmitted disease; STI: sexually transmitted infection.

p < .01; *p < .05; †p < .10.

First, not surprisingly, prior testing status predicted subsequent testing status, β = .84, t = 13.75, p < .01. Second, the main effect model did not account for an additional proportion of the variance in testing status, R²Δ = .01, F(4, 74) = .60, p > .10, as all predictors were nonsignificant (all βs < |.06|, ts < .52, ps > .10). Third, although the overall two-way interaction model did not account for additional variance, R²Δ = .02, F(6, 74) = 1.06, p > .10, we had a priori interest in examining the three interactions involving handedness strength. Indeed, as expected, the comparative risk × handedness strength interaction was significant (β = −.16, t = −2.00, p < .05). Moreover, the absolute verbal risk × handedness strength interaction was marginally significant (β = .27, t = 1.77, p = .08). The absolute numeric risk × handedness strength interaction was not significant (β = −.09, t = −.70, p > .10; all other two-way interactions were not significant, βs < |.07|, ts < .50, ps > .10). To interpret the significant (and marginally significant) interactions, testing status was graphed as a function of handedness strength at 1 SD above and below the mean of risk estimates (see Figure 2). First, testing status appeared more impacted by social comparative risk estimates in MH, where higher comparative risk estimates were associated with testing. Second, testing status in SH showed the opposite pattern, where feeling at higher absolute risk was associated with testing. Interestingly, MH appeared to show the opposite pattern from SH here, where reports of being at higher absolute risk were associated with less testing.²

Figure 2.

Testing status as a function of handedness strength, absolute verbal risk, and comparative risk. Note: STD/STI testing status was measured using a dichotomous question (0 = no; 1 = yes).

Discussion

Comparative and absolute risk perceptions relate to affective, cognitive, and behavioral responses to health threats. The current research provides evidence that handedness strength—a proxy for individual differences in interhemispheric interaction—moderates the association between risk perceptions and worry. Specifically, comparative risk was correlated with worry among MH but not SH, whereas absolute risk was more correlated with worry for SH than for MH. Moreover, we found some preliminary evidence that an adaptive behavioral response to feeling at risk for an STD/STI—getting tested—appeared to show a similar pattern.

Limitations and future directions

There are several limitations to the current study that warrant exploration in future research. First, we examined our core hypotheses using a single domain. As stated previously, we chose STD/STI risk because it was deemed to be more relevant to our population than traditional events and has social implications. Although we suggest that our results would replicate in other domains, future research is needed to test such assumptions.

Second, although our theoretical account for the results involves individual differences in organizational brain structure and connectivity, we have no direct evidence of these mechanisms. Our use of handedness strength as a proxy variable is clearly a limitation. Although research has illustrated a link between handedness and callosal size, future research should provide more definitive evidence that underlying brain structures and processes are the critical for our results.

Third, we have suggested that worry is an affective outcome that is predicted by risk perceptions. However, given that we did not manipulate risk perceptions, we cannot draw any causal conclusions. Although some researchers have articulated that worry can be a precursor to adaptive cognitions and behaviors (Davey, 1993; Leventhal et al., 2003; Miller et al., 1996), other researchers have suggested the opposite pathway (Loewenstein et al., 2001; see also Betsch et al., 2010) or have treated worry as a moderator variable (Klein et al., 2009). Although our research examined the effect of risk perceptions on a behavioral measure, future research is needed to gain a deeper understanding of the relationship between risk, handedness, and other outcomes.

Conclusions

Risk is an important component of health behavior models; yet we have an incomplete picture of when comparative versus absolute risk will predict cognition, affect, and behavior. The current research suggests that a fundamental difference in brain hemisphere organization is an important moderator, which could help account for mixed results. For instance, if a particular study had more MH (vs SH), this could account for the results showing that participants were more sensitive to comparative (vs absolute) risk information. Moreover, these results suggest that it may be prudent to tailor risk communications to the characteristics of the individual in order to achieve maximum efficacy in shaping affective, cognitive, and behavioral outcomes.

Footnotes

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Notes

References

Betsch

Renkewitz

Betsch

. (2010) The influence of vaccine-critical websites on perceiving vaccination risks. Journal of Health Psychology 15: 446–455.

Blalock

DeVellis

Afifi

. (1990) Risk perceptions and participation in colorectal cancer screening. Health Psychology 9: 792–806.

Cabeza

Nyberg

(2000) Imaging cognition II. An empirical review of 275 PET and FMRI studies. Journal of Cognitive Neuroscience 12: 1–47.

Chapman

Coups

(2006) Emotions and preventative health behavior: Worry, regret and influenza vaccination. Health Psychology 25: 82–90.

Christman

Jasper

Sontam

. (2007) Individual differences in risk perception versus risk taking: Handedness and interhemispheric interaction. Brain and Cognition 63: 51–58.

Claassen

Henneman

Kindt

. (2010) Perceived risk and representations of cardiovascular disease and preventive behaviour in people diagnosed with familial hypercholesterolemia: A cross-sectional questionnaire study. Journal of Health Psychology 15: 33–43.

Clarke

Zaidel

(1994) Anatomical-behavioral relationships: Corpus callosum morphometry and hemispheric specialization. Behavioural Brain Research 64: 185–202.

Davey

GCL

(1993) A comparison of three cognitive appraisal strategies: The role of threat in devaluation of problem-focused coping. Personality and Individual Differences 14: 535–546.

Dillard

Ferrer

Ubel

. (2012) Risk perception measures’ associations with behavior intentions, affect, and cognition following colon cancer screening messages. Health Psychology 31: 106-113.

10.

Feinberg

Keenan

. (2005)Where in the brain is the self? Consciousness and Cognition 14, 661–678.

11.

French

Hevey

Sutton

. (2006) Personal and social comparison information about health risk: Reaction to information and information search. Journal of Health Psychology 11: 497–510.

12.

French

Sutton

Marteau

. (2004) The impact of personal and social comparison information about health risk. British Journal of Health Psychology 9: 187–200.

13.

Happe

Brownell

Winner

(1999) Acquired “theory of mind” impairments following stroke. Cognition 70: 211–240.

14.

Harris

Sparks

Raats

(2002) Theoretical and applied issues in the provision of absolute and comparative risk information. Risk, Decision and Policy 7: 153–163.

15.

Harris

Smith

(2005) When the risks are low: The impact of absolute and comparative information on disturbance and understanding in US and UK samples. Psychology & Health 20: 319–330.

16.

Janz

Becker

(1984) The health belief model: A decade later. Health Education Quarterly 11: 1–47.

17.

Jasper

Christman

(2005) A neuropsychological dimension for anchoring effects. Journal of Behavioral Decision Making 18: 343–369.

18.

Jasper

Barry

Christman

(2008) Individual differences in counterfactual production. Personality and Individual Differences 45: 488–492.

19.

Klein

WMP

(1997) Objective standards are not enough: Affective, self-evaluative, and behavioral responses to social comparisons information. Journal of Personality and Social Psychology 72: 763–774.

20.

Klein

WMP

(2002) Comparative risk estimates relative to the average peer predict behavioral intentions and concern about absolute risk. Risk, Decision and Policy 7: 193–202.

21.

Klein

WMP

Zajac

Monin

(2009) Worry as a moderator of the relationship between risk perceptions and quitting intentions in adult and young adult smokers. Annals of Behavioral Medicine 38: 256–261.

22.

Leventhal

Brissette

Leventhal

(2003) The common-sense model of self-regulation of health and illness. In: Cameron

Leventhal

(eds) The Self-Regulation of Health and Illness Behaviour. London: Routledge, pp. 42–65.

23.

Lipkus

Kuchibhatla

McBride

. (2000) Relationship among breast cancer perceived absolute risk, comparative risk, and worries. Cancer Epidemiology Biomarkers & Prevention 9: 973–975.

24.

Loewenstein

Weber

Hsee

. (2001) Risk as feelings. Psychological Bulletin 127: 267–286.

25.

McCaul

Mullens

(2003) Affect, thought, and self-protective health behavior: The case of worry and cancer screening. In: Suls

Wallston

(eds) Social Psychological Foundations of Health and Illness. Malden, MA: Wiley-Blackwell. 137–168.

26.

Mason

Prevost

Sutton

(2008) Perceptions of absolute versus relative differences between personal and comparison health risk. Health Psychology 27: 87–92.

27.

Miller

Shoda

Hurley

(1996) Applying cognitive-social theory to health protective behavior: Breast self-examination in cancer screening. Psychological Bulletin 119: 70–94.

28.

Oldfield

(1971) The assessment of handedness: The Edinburgh inventory. Neuropsychologia 9: 97–113.

29.

Perry

Rosen

Kramer

. (2001) Hemispheric dominance for emotions, empathy, and social behavior: Evidence from right and left handers with frontotemporal dementia. Neurocase 7: 145–160.

30.

Propper

Christman

Phaneuf

(2005) A mixed-handed advantage in episodic memory: A possible role of interhemispheric interaction. Memory & Cognition 33: 751–757.

31.

Rankin

Gorno-Tempini

Allison

. (2006) Structural anatomy of empathy in neurodegenerative disease. Brain 129: 2945–2956.

32.

Rose

(2010) Are direct or indirect measures of comparative risk better predictors of concern and behavioral intentions? Psychology and Health 25: 149–165.

33.

Rose

(in press) Debiasing comparative optimism and increasing worry for health outcomes. Journal of Health Psychology.

34.

Rose

Jasper

Corser

(2012) Interhemispheric interaction and egocentrism: The role of handedness in social comparative judgement. British Journal of Social Psychology 51: 111–129.

35.

Ruby

Decety

(2004) How you feel versus how do you think she would feel? A neuroimaging study of perspective-taking with social emotions. Journal of Cognitive Neuroscience 16: 988–999.

36.

Witelson

Goldsmith

(1991) The relationship of hand preference to anatomy of the corpus callosum in men. Brain Research 545: 175–182.

37.

Zajac

Klein

McCaul

(2006) Absolute and comparative risk perceptions as predictors of cancer worry: Moderating effects of gender and psychological distress. Journal of Health Communication 11: 37–49.