Anxiety, Depression, and Goal-Seeking in Conservatives, Liberals, and Moderates

C. Brack and X. Zhang


Anxiety and depression would not seem to be specifically relevant to a particular individual's political affiliation. However, what should ostensibly be non-correlated variables have turned out to be highly correlated in our March 2005 survey. We found a statistically significant relationship (p < .0001) between anxiety, depression, and liberalism.

Before we try to answer why, let's go over the data, and then review the curious relationship between anxiety and depression and some interesting variations in adaptive behaviors that conservatives, liberals, and moderates engage in.

Anxiety is a generic term used to summarize a variety of psychological illnesses such as social anxiety disorder, agoraphobia, panic syndrome, post-traumatic stress disorder, generalized anxiety syndrome, OCD, and other hyperphobic disorders. In this survey, we only were able to trap chronic generalized anxiety.

Depression is a generic term used to summarize a variety of psychological illnesses such as major depressive disorder, dysthymia, bipolar disorder, cyclothemia, Seasonal Affective Disorder (SAD), and postpartum depression. We did not trap specific depressive disorders, except in the case of bipolar disorder.

Anxiety and depressive disorders share common neurological substrates, and can be co-classified into three categories: somatic anxiety (panic type disorders), anhedonic depression (lack of positive feelings and withdrawal from goal-seeking), and general distress (exhibited by all depressive and anxiety disorders).

Theories of anxiety and depression vary by type, and have been proposed to involve a variety of genetic, environmental, anatomical, pathological, and neurochemical hypotheses. Regardless of the cause, treatments are most frequently targeted to neurotransmitter regulation. The monoamine hypothesis has been a popular but incomplete model for depression, as it does not explain the non-impact of monoamine depletion in normal control subjects.


Survey Data

The survey consisted of 1,390 people, of which 615 were conservative, 569 were liberal, and 206 were moderate. Figure 1 contains the percent of those reporting chronic anxiety by political and gender cohorts. Note that L=Liberal, M=Moderate, and C=Conservative will be used on all graphs.

I. Anxiety and Depression Levels by Political and Gender Cohorts.

Figure 1: Chronic Anxiety Levels by Political and Gender Cohorts

The elevation in reported anxiety for liberals is statistically significant (p < .01). Moderates report slightly higher anxiety ratings than conservatives. The male and female percents are consistent within political cohort, with males reporting slightly higher anxiety on average.


Figure 2: Depression Levels by Political and Gender Cohorts.

Another statistically significant (p < .01) elevation in the reported rate of depression for liberals, 34.3%, which is slightly less than their reported anxiety rating of 35.9%. The conservative depression rate of 24.7% closely shadows their anxiety rate of 24.0%. Moderates drop to a 21.8% depression rate, down from their 25.7% anxiety rate. Variances by gender are greater in anxiety than in depression.

Comorbidity of anxiety and depression includes only those reporting "yes" to both anxiety and depressive symptoms. Only 10.7% of conservatives reported anxious and depressive symptoms, which compares to 12.6% for moderates, and 18.6% for liberals. By gender, the conservative females report combined anxiety-depression symptoms at a rate of 14.9%, liberal females at 21.4%, and moderate females at 17.8%. The male conservatives report 9.8% anxiety-depression comorbidity, male liberals at 16.8%, and male moderates at 10.4% (see Figure 3).


Figure 3: Comorbidity of Anxiety-Depression Levels by Political and Gender Cohorts.

II. Goal-Seeking and Multi-tasking by Political and Gender Cohorts

As we see from the responses to the goal-seeking question in our survey: "If you play, you should play to win", we see an interesting variation in responses across political affiliation:

Figure 4: Goal-seeking by Political and Gender Cohorts

This question is correlated with the psychological concept of "goal-seeking", but still only a proxy for that complex behavior. Still, we find statistically significant (p < .01) variations in our political cohorts, with the highest positive response rates among conservatives (72.7%), lowest among liberals (38.7%), and our moderates once again fall in between at (60.7%). If we look at the anxiety-depression breakdown by political cohort, we see an interesting impact on goal-seeking (Figure 5).

Figure 5: Impact of Anxiety and Depression on Goal-seeking by Political Cohorts (A+D=Anxiety+Depression, A=Anxiety only, D=Depression only, N=Normal)

In Figure 5, we see the lowest rates of goal-seeking by political cohort are in the depression-only category, and consistent with a primary symptom of anhedonic depression. Conversely, the anxiety-only groups have the highest rate of goal-seeking across all political cohorts, even exceeding the normals (people that reported no anxiety or depression). Comorbid anxiety-depression was higher than depression-only in all political cohorts, but not as high as the anxiety-only group.

While anxiety does not reduce the level of goal-seeking as compared to normals, it does tend to change the way that the goal-seeking behavior is executed. When asked the question "I frequently stop in the middle of a task and start a new one", liberals answered "yes" to this question 52.9% of the time, and conservatives 41.6%. Once again, moderates fell in between at 47.6% (Figure 6).

Figure 6
: Multi-tasking propensities by Political and Gender Cohorts


If we look at the anxiety-depression breakdown by political cohort, Figure 7, we see a significant elevation in multitasking with the anxiety-only and comorbid anxiety-depression categories as compared with normals. Anxiety correlates highly with a multi-tasking behavioral profile. This is not without precedent in the animal kingdom, as this same type of anxious multi-tasking is seen in animals that live in predator-rich environments.

Figure 7
: Impact of Anxiety and Depression on Multi-tasking by Political Cohorts (A+D=Anxiety+Depression, A=Anxiety only, D=Depression only, N=Normal)

The depression-only groups have a similar but generally smaller elevation in multi-tasking behavior than normals

III. Offspring Entitlement by Political and Gender Cohorts

We also find an interesting correlation between anxiety and offspring emtitlement. We asked the 1,390 participants of this survey if they "wanted to leave a lot of money to their kids when they die". This question is not very useful as an indicator of offspring entitlement, due primarily to the lack of controls for wealth and whether or not the subject had children. However, it is useful to examine the impacts of anxiety and depression on offspring entitlement. If we look at the breakdown by political and gender cohorts, Figure 8, we see liberals answering "yes" at 40.6%, moderates at 53.9%, and conservatives at 60.5%.

Figure 8
: Offspring Entitlement by Political and Gender Cohorts

We can see the impact of anxiety and depression on offspring entitlement in Figure 9.

Figure 9
: Impact of Anxiety and Depression on Offspring Entitlement by Political Cohorts (A+D=Anxiety+Depression, A=Anxiety only, D=Depression only, N=Normal)

We see elevations in response rates for both the anxiety-only and comorbid anxiety-depression groups across all the political cohorts as compared to the normals. The depression-only groups have a smaller elevation in the conservative and liberal cohorts. We do not find this trend in moderates, but this may be due to their low sample size (n=206).

IV. Language Comprehension by Political and Gender Cohorts

If we examine the problems in language comprehension by political affiliation, Figure 10, we see the predicted elevation in problems with the liberal, 44.8%, relative to both the conservative, 37.4%, and moderate at 42.7%. Males tend to report higher language comprehension problems.

Figure 10: Language Comprehension Problems by Political and Gender Cohorts

The impact of anxiety and depression is dramatic across all political cohorts (Figure 11). Anxiety and depression correlate highly with conversational language comprehension problems.

Figure 11
: Impact of Anxiety and Depression on Language Comprehension by Political Cohorts (A+D=Anxiety+Depression, A=Anxiety only, D=Depression only, N=Normal)

IV. Word Pair Test by Political and Gender Cohorts

In the results from our Word Pair Test, where we ask the subject to give a preference for a semantic versus shape translation for a word pair, we find an elevation with conservative males in the preference for the semantic translation (Figure 12).

Figure 12
: Preference for Semantic vs. Shape Translation by Political and Gender Cohorts

The impact of anxiety and depression on semantic preference is exhibited in figure 13.

Figure 13
: Impact of Anxiety and Depression on Semantic vs. Shape Translation by Political Cohorts (A+D=Anxiety+Depression, A=Anxiety only, D=Depression only, N=Normal)

Among liberals and moderates, the anxiety-only and comorbid anxiety-depression groups exhibited a lesser preference for semantic translation of the word pair than normals in their respective political cohorts. There was no corresponding trend with conservatives.

V. Discussion

The correlation between anxiety, depression, and liberalism poses several interesting questions. The first and most obvious is---why? In our previous research (1), we had detected a rightward bias in hemisphericity of the liberal as compared to the leftward bias in the conservative. There has been considerable research on the emotional attributes of the two hemispheres, and the right hemisphere has been implicated by many studies in negative emotions and depressive moods (2) (see Borod et al (3) for a more recent review of the research). Electroconvulsive therapy is more effective when applied to the right hemisphere than the left. (4)

The right hemisphere has also been implicated in a variety of anxious illnesses, but the research is equivocal. Brain tumors in the right hemisphere have been identified in anxiety disorders by Mainio et a (5), whereas left hemispheric tumors seem to have little or no impact. Hughes et al (6) found a strong link between norepinephrine levels in urine during depressive and anxious episodes, further implicating the right hemisphere's noradrenergic bias in mood disorders. Wittling (7) proposes that the right hemisphere is dominant in modulating the sympathetic activity of the heart, regulating both strength and frequency of the heart beat. This is further evidence of a strong connection between the right hemisphere, anxiety, and the sympathetic nervous system.

The correlation between anxiety, depression, and liberalism is not surprising given their common right hemispheric bias. Indeed, this correlation is further evidence of an elevated right hemispheric orientation of the liberal, when compared to the conservatives and moderates.

Further, the strong correlation between goal-seeking and conservatism naturally raises the question of the role of dopamine, of which considerable evidence exists for mediating goal-seeking behavior (8). Tucker and Williamson (9) put forth the theory of an asymmetric distribution of dopamine and norepinephrine in the two hemispheres of the brain, with the left hemisphere organized around a dopaminergic activation system, and the right hemisphere organized around a noradrenergic activation system.

The goal-seeking orientation of conservatives is consistent with a left hemispheric bias and corresponding bias towards dopaminergic behavioral attributes, which may also be related to the elevation in the conservative's binary reality-model preference noted by Brack and Zhang. Binary reality-models are conducive to goal-seeking, as they organize the analysis of stimuli into reward and punishment categorizations from which goal-seeking behavior can be mediated.

The survival value of anxiety is obvious---it prepares the individual for potential threats. The survival value of depression is not that obvious, and a variety of theories, such as the Watson and Andrews "social navigation" hypothesis (10) proposes, among other things, that depression promotes niche changes when the current niche is restricting, thereby improving overall adaptability.

We may have found evidence of a divergence in the types of survival approaches that conservatives and liberals engage in, and that divergence may be connected to the susceptibility of liberals to anxiety and depression. Liberals show a greater tendency for multi-tasking, which is also a trait of people reporting depression and chronic anxiety. In certain threatening conditions, a propensity towards multi-tasking would have a positive survival value over single-tasking.

We see a strong elevation in goal-seeking behavior in the conservatives, which is an indication that conservatives may be more "active" in their immediate environments, at least when it comes to "rewarded" behaviors. We also found that chronic anxiety is also highly linked to goal-seeking and "rewarded" behaviors. Chronic anxiety may be the right hemisphere's answer to the left-hemisphere's propensity for goal-seeking. In other words, chronic anxiety may be a hyper-activation of the right hemisphere that promotes goal-seeking when the left hemisphere's dopaminergic system is not effective. However, the direction of causality is not clear, and an individual forced into "goal-seeking" might develop chronic-anxiety, rather than the other way around. However, a bidirectional causal relationship is also possible.

Also noteworthy is the phenomenon of "offspring entitlement", or the desire to improve the survivability of offspring. Anxious and depressive subjects had a noted elevation in offspring entitlement when compared to normals.

Liberals report elevated problems with language comprehension, but the impact of depression and anxiety on language comprehension is severe across all political cohorts. Depression and especially anxiety dramatically impair the ability to track and maintain focus during conversation. This is consistent with other research (11), and also consistent with the right hemisphere's general deficit for literal language processing. The survival value of this language impairment is simply in the redirection of mental resources from the conversation at hand, and to potential threats in the environment.

The questions can now be asked: are conservatives and liberals complementary in their survival behaviors? Does a goal-seeking conservative fit in better with a non-goal seeking liberal? Does the single-tasking conservative fit better with the multi-tasking liberal? Does an optimal combination of conservatives, moderates, and liberals fill in the various socioeconomic niches better than all conservatives or all liberals or all moderates? Does society perform "better" with an optimal mix of conservatives, liberals and moderates? Obviously, these questions will not be easily or unequivocally answered.



1. Brack and Zhang. Right and Left Hemispheric Biases in Political and Religious Tendencies: Part 1., March 2005.

2. Schwartz, JA, Speed NM, Brunberg JA, Brewer TL, Brown M, Greden JF. Depression in Stroke Rehabilitation. Biological Psychiatry. 1993 May 15;33 (10):694-9.

3. Borod JC, Bloom RL, Brickman AM, Nakhutina L, Curko EA. Emotional Processing Deficits in Individuals with Unilateral Brain Damage. Applied Neuropsychology, 2002;9(1):23-36.

4. Springer, S.; Deutsch, G. Left brain, right brain. New York: W. H. Freeman and Co.; 1985

5. Mainio A, Hakko H, Niemela A, Tuurinkoski T, Koivukangas J, Rasanen P. The Effect of Brain Tumour Laterality on Anxiety Levels Among Neurosurgical patients. Journal of Neurology, Neurosurgery, and Psychiatry 2003 Sep;74 (9):1278-82.

6. Hughes JW, Watkins L, Blumenthal JA, Kuhn C, Sherwood A. Depression and anxiety symptoms are related to increased 24-hour urinary norepinephrine excretion among healthy middle-aged women. Journal of Psychosomatic Research. . 2004 Oct;57(4):353-8.

7. Wittling W, Block A, Genzel S, Schweiger E. Hemisphere asymmetry in parasympathetic control of the heart. Brain Cogn. 1998 Oct;38(1):17-35.

8. Irene A. Yun, Ken T. Wakabayashi, Howard L. Fields, and Saleem M. Nicola. The Ventral Tegmental Area Is Required for the Behavioral and Nucleus Accumbens Neuronal Firing Responses to Incentive Cues.

The Journal of Neuroscience, March 24, 2004, 24(12):2923-2933.

9. Tucker and Williamson. Asymmetric Neural Control Systems in Human Self-Regulation. Psychological Review, 1984, 91, 185-215.

10. Watson, PJ, Andrews PW. Toward a revised evolutionary adaptationist analysis of depression: The social navigation hypotheis. Journal of Affective Disorders 72, 1-14.

11. Hopko DR, McNeil DW, Lejuez CW, Ashcraft MH, Eifert GH, Riel J. The effects of anxious responding on mental arithmetic and lexical decision task performance. Journal of Anxiety Disorders 2003;17(6):647-65