Thursday, April 28, 2011

When Would We Predict Change Blindness Among Conservatives and Liberals?

In Jost, Glaser, Kruglanski, and Sulloway, 2003 the authors describe a number of possible relationships that can transform an individual’s perception of environmental uncertainty and environmental threat into political conservatism.  The authors characterize political conservatism as comprised of both resistance to change and the endorsement of social and economic inequalities.  Jost et al. demonstrate that the relationships between uncertainty about one’s environment and feeling threatened by one’s environment and political conservatism are complex and vary by individual.  These relationships include the moderating roles of:
  • Epistemic motives: including dogmatism, intolerance of ambiguity, uncertainty avoidance, and the needs for order, structure, and closure.
  • Existential motives: including self-esteem, a  prevention-oriented regulatory focus, and terror management.
  • Ideological motives: including the rationalization of self-interest, group-based dominance, and system justification.

In addition, the particular form that political conservatism takes reflects socially-learned ideology-influenced knowledge, including beliefs about cosmology, human nature, governance, etc.   

The authors argue that while both directed reasoning—reasoning in pursuit of a certain belief and undirected reasoning—reasoning by a process that satisfies meta-cognitive needs—influence the acquisition of new knowledge, prior knowledge influences when and how these needs are met.  For example, an individual may believe that the social world is dangerous because they were taught that it is dangerous or have experienced it as dangerous, but they may also, as Duckitt, Birum, Wagner, & du Plessis, 2002 demonstrated using structural equation modeling, tend to believe that the social world is dangerous as a consequence of a need for social conformity.  In other words, the need for social conformity, once formed as a personality variable, may motivate the acquisition of evidence in favor of believing that the world is a dangerous place.   Duckitt et al’s study was not longitudinal and could not study reciprocal relationships between social conformity needs and belief in a dangerous world over time. 

This potentially reciprocal relationship between personality traits, i.e. core motives, and ideology-relevant knowledge is particularly interesting when one looks at attitudes towards change.  Change in general, Jost et al. argue, tends to lead conservatives to feel threatened.  Instead, conservatives tend to positively value stability.  At the same time, political conservatives will often advocate a return to an idealized past, both more recent and more distant, denigrating the present as, to cite adjectives from the Right-Wing Authoritarianism scale, “radical” (in itself a negative adjective), “perverse,” or “sinful.”   This surface-level contradiction could be resolved if we more carefully defined what kinds of changes conservatives are threatened by, across a variety of situations. 

Reasoning by analogy to studies of change-blindness attention and perception, I wonder if the changes that conservatives have such negative affective reactions to are schema-threatening changes.  Schemas, mental models that are used to encode, store, and recall information, are relatively stable tools which have the side-effect of limiting perception of variation in schema-consistent information.  In other words, if change in the environment does not challenge the schema, it is often ignored completely.  Further, if a change would challenge the schema, it is often ignored if most of the information in the environment supports the schema.  For example, I myself did not notice that parts of a color photograph were printed in black and white even though that photograph was presented in an article on change blindness. 

Change-blindness studies are often even more dramatic and of direct-relevance to interpersonal interaction.  In one study, an experimenter sent a confederate out on a college campus to ask people for directions.  After the experimenter greeted the subject, another set of confederates walked between the dyad while carrying a large object that completely obscured the subject’s view.  In the meantime, another confederate replaced the first confederate, so that the subject was now giving directions to a completely different person.  Many subjects failed to notice this change, even when the person was much shorter, taller, of a different skin-tone, etc.   Further studies revealed certain patterns—college student subjects were more likely to notice that the college-student aged confederate had changed, changes in gender (if I recall) were noticed, as, sometimes, were changes in race (Simons & Levin, 1997).

If change blindness is a demonstrable phenomenon in such a basic social situation, it may not be surprising if it occurred in relation to ideological schema as well, allowing for even some schema-inconsistent changes to go unnoticed and thus never become threatening.  The question then arises, what predicts when a schema-inconsistent change will be noticed and how threatening it will become?  Further, when a country’s politics are polarized around a left-right dimension, are there concrete and complex schemas that identify any action of the left as being in favor of “change.”  Are these two types of change—change that is noticed because a stimulus is schema-inconsistent and change that is noticed because a stimulus is consistent with a schema that labels that stimulus as change—correlated to different affective reactions in conservative individuals?  Or does it all come out in the wash?  

Ideology as Variation in Attitudes Towards Uncertainty and Threat (in the US, Europe, Israel, and New Zealand)

Ideology: Its Resurgence in Social, Personality, and Political Psychology
To request see:

Jost, Nosek, and Gosling argue in 2008 that ideology comprises a “single left–right (or liberal–conservative) dimension” that “possesses two core aspects: (a) advocating versus resisting social change and (b) rejecting versus accepting inequality” (Jost, Nosek, & Gosling, 2008). They further argue that “ideological belief systems may be structured according to a left–right dimension for largely psychological reasons linked to variability in the needs to reduce uncertainty and threat.” (Jost, Nosek, & Gosling, 2008) In these statements, Jost, Nosek, and Gosling do not explicitly discuss causal relationships between ideology and personal “needs.” Indeed, they are explicitly, at least for the purposes of this paper, agnostic as to whether “left–right proclivities stem from basic, underlying preferences that are apolitical in nature or, alternatively, that the adoption of specific ideologies leads people to internalize a host of extremely general attitudes concerning stability versus change and hierarchy versus equality” (Jost, Nosek, & Gosling, 2008). Is each line of argument that Jost et al. employ in defense of their thesis also agnostic as to the causal role of “underlying preferences”?

Their first set of arguments concerns implicit ideological preferences. For example, they specifically examine preferences for “tradition, conformity, order, stability, traditional values, and hierarchy—versus those for progress, rebelliousness, chaos, flexibility, feminism, and equality” (Jost, Nosek, & Goslking, 2008). These preferences, Jost et al. state, fall into two negatively correlated groups of preferences. Any one person would be somewhat unlikely to have strong implicit preferences for, to take an example, tradition and stability as well as progress and chaos. In addition, any one person who has a preference for tradition is somewhat more likely to have a preference for stability as well. These preferences do correlate to political orientation, but Jost et al. are agnostic as to the direction of causation, but do suggest that the negative correlation between preferences is a “constraint” on the development of political orientation. How strong the constraint is would depend on the strength of the negative correlation and individual factors.

Their second set of arguments concerns system justification. System justification theory does make a specific causal argument—mainly that the need to believe in a just world does cause more automatic and more deliberate system-justifying behaviors, as Jost and colleagues have defined and measured them. However, the strength of this need differs both between individuals and within individuals. Prolonged threat, for example, can increase system-justifying behaviors. The need to believe in a just world also correlates to variables used to measure political-orientation. In this argument, political upbringing may influence the strength of one’s need to believe in a just world, which in turn would influence the way that one comes to believe in a just world—by, for example, justifying different aspects of the existing economic-social-political system and not justifying others.

Their 3rd and 4th sets of arguments concern individual preferences. According to Jost et al., conservative and liberal orientations are correlated to different “big five” personality traits as well as to different self-reported preferences for attitude objects like fishing and one’s father versus atheists and tattoos. These lines of argument are open to the same question of causality. Are certain personality traits expected of individuals raised in social contexts that teach and expect a certain political orientation? Or, alternatively, are they more “inborn” with these inborn traits reflected in political orientation? To address the fourth line of argument, are certain lifestyle preferences caused by certain individual fundamental preferences or are they merely artifacts of cultural differences in upbringing between liberals and conservatives?

Jost et al.’s fifth line of argument is central to their thesis that the need to manage uncertainty and threat is central to political orientation. Here they present two main arguments. The first is that a conservative shift corresponds to increased exposure to both threats to the system and mortality salience—as demonstrated both by in-laboratory reminders of 9/11, personal experience of terrorist attacks (both in the US and in Europe) and their effects, and rapid “social change and the fear of social decline.” The second is that liberal shifts tend to be caused by having to approach positive novel experiences and perspectives. These two sets of arguments differ, of course, in degree of control and in the nature of threat. In the conservative shift, direct control is extremely low, in the liberal shift, direct control is relatively high. Personal threat is relatively high in the conservative shift case and relatively low in the liberal shift case. In either case, there’s some evidence that the precipitating events (attack, novel experiences) changed personal and political preferences.

How would you tell the causal story? What creates consistency between political orientation and other preferences and what could disrupt this consistency?