Will We Too Become Our Racist Grandparents?

A body of social scientific research supports the commonly-observed phenomenon of older people being more prejudiced/racist than younger people, and it is often assumed that this difference reflects changing social values toward egalitarianism and acceptance of diversity.

However, my recent reading of the role of automatic and control processes in attitudes has made me question the validity of that assumption.

In very simplistic terms, various social cognitive models suggest that two different (but interacting) mental processes are involved in determining our attitudes. The automatic process involves the activation of a network of associations that have been created by conditioning and is evoked by stimuli without a person's intent (and, sometimes, without awareness). The associative evaluations that arise from this automatic process (aka implicit attitudes) do not necessarily reflect what we would endorse as true. For instance, a person might have a positive implicit attitude for Bud Light due to exposure at a vulnerable age to Spuds McKenzie TV advertisements (a dog on the beach in sunglasses, yay) while not believing that it is in any way true that Bud Light is a good beer or at all a desirable product. Less trivially, a person may (indeed, probably does) have negative associations with black people and positive associations with white people; indeed, black people themselves may have these associations, if to a lesser extent than white people. (Do you doubt that you have this kind of implicit prejudice? Feel free to put your implicit moral superiority to the test on the Implicit Association Test.)

The control stage is a conscious, effortful process of placing one's automatic evaluations under judgment to determine what a person believes is true. For example, one can consider the Bud Light-Spuds McKenzie pairing rationally in the light of other propositional beliefs such as "Spuds McKenzie doesn't even drink beer" and "Bud Light tastes bad" to develop a negative explicit attitude toward Bud Light (e.g. "I don't like Bud Light") that diverges from the association-based response and to behave in keeping with that explicit attitude by not purchasing it. Similarly, a person may have negative associations with black people while recognizing that black people are not actually inferior and not behaving in a prejudicial manner toward them.

But it's important to note that these models often (always? I'm not familiar with all the variants on the theme) state that the default situation is for the control process to merely rubber-stamp the attitude activated by the automatic process rather than place it under rational scrutiny. Therefore, without the consideration of other beliefs, these automatic evaluative associations can become one's attitude, full-stop.

But if the control process is effortful (draws on a cognitive resource), I thought: shouldn't older people, who have deterioration in cognitive resources, be less well-equipped to over-ride the racist attitudes popping up from their automatic process? Could this contribute to the fact that older people are more prone to behave in racist ways? (I have to credit a paper I read for cognitive psychology on the greater difficulty older people have in inhibiting their responses to distracting visual stimuli in a perception task for giving me the idea that they might have difficulty inhibiting their responses to automatically activated associations also.)

Of course, as with a dozen other good ideas I've had on related topics today, I have been beaten to the punch on this question. In a study published in March of this year, an analysis of almost 16,000 respondents on the IAT web site provided support of their (and my) hypothesis that older adults show greater bias against blacks and greater preference for whites on the IAT compared to younger people because older people are less able to control their automatic associations, not because their automatic associations are more extreme (Gonsalkorale, et al 2009).

But wait a minute...I thought that you said that the IAT measures implicit attitudes, which is what automatic evaluative associations are. Now you're saying that people control or regulate those automatic associations in the process that is used to measure the very same associations. That's confusing. Well, yeah, it sort of is, and is a good example of how we need to be careful of equating a process with a measurement of the results of the process. The researchers state: "However, although implicit measures surely restrict the role of inhibitory processes more so than do self-report measures, even here, self-regulatory abilities affect task performance."

The implications of this research are kind of depressing: those of us younger folk who listen with embarrassment and dismay as our elders make weird racist comments may not actually be less implicitly racist than they are, but only better able to keep those (untrue) associations at bay through the application of cognitive effort. Once we're old fogeys ourselves, we may be just as apt to say and do racist things.

To me, this points up the importance of people developing less prejudicial evaluative associations in the first place or having patterns of associative activation that are less negative toward racial or ethnic minority groups (since different stimuli or contexts can cause different parts of your associations to be activated - like thinking of black men when seeing a photo of Denzel Washington vs. a gang member). Is it completely naive to hope that the next generation, growing up in a world in which the elementary school walls will show a line of US Presidents that includes a black man, may be less implicitly racist than our generations are?

A literature search reveals one tiny glimmer of hope in this regard. A July 2009 study found lower levels of anti-black implicit attitudes among non-black college students at the time of the Obama presidential campaign than before, particularly among those who listed a positive role model (e.g. Obama, MLK Jr.) among their list of 5 thoughts that came to mind about black people (measured after the IAT) (Plant et al, 2009). Of course, it is very possible that the various positive Obama phenomena will not yield persistent changes in attitudes, implicit or explicit.

Sources:

Gonsalkorale K, Sherman JW, and Klauer KC (2009). Aging and prejudice: Diminished regulation of automatic race bias among older adults, Journal of Experimental Social Psychology, 45, 410-414.

Plant EA, Devine PG, Cox WTL, et al (2009). The Obama effect: decreasing implicit prejudice and stereotyping, Journal of Experimental Social Psychology, 45(4), 961-964.