It is for instance the case of the experiment of Shayo & Harel 2011 who point out that a significant part of the constituency is always motivated by non-instrumental motives and that, intuitively, that part increases as much as decreases the probability of being pivotal. With the help of a different experiment, Kamenica & Egan 2012 demonstrate that when the poor vote against redistribution policies they are motivated by expressing an ideology and the probability of being pivotal (and, consequently, of gaining from the policy) is largely irrelevant because, even if positive, it is overcome by such expressive motives.
Passing to the first attempts to apply neuroscientific methods to voting, Rule et al. 2010 observed higher activations in the amygdala of people in the US and in Japan when facing the politicians they have chosen to vote. Furthermore, Kanai et al. 2011 show that political orientations correlate with different brain structures (liberalism implies bigger grey matter volume in the anterior cingulate cortex, conservatism correlates with more volume in right amygdala). Bischoff et al. 2012, finally, found neural activations confirming the importance of expressive motives in particular for people classified as non-altruistic.
In summary, these works suggest also an explanation for the recent success gained by sentiment analysis of messages published on social media before elections. If people vote mainly because of the desire for self-expression, why not using an amplifier such as Facebook or Twitter?
Shayo & Harel (2011), "Non-consequentialist voting", Journal of Economic Behavior & Organization, 81(1): 299-313
Kamenica & Egan (2012), "Voters, Dictators, and Peons: Expressive Voting and Pivotality", Public Choice, forthcoming
Rule et al. (2010), "Voting behavior is reflected in amygdala response across cultures", Soc Cogn Affect Neurosci, 5 (2-3): 349-355
Kanai et al. (2011), "Political Orientations Are Correlated with Brain Structure in Young Adults", Curr Biol. 2011 April 26; 21(8): 677-680
Bischoff et al. (2012), "The Neuroeconomics of Voting: Neural Evidence of Different Sources of Utility in Voting", MAGKS Papers on Economics, n. 201234