Verified by 
In the aftermath of a week of terror in Boston, we still know little about what led the Tsarnaev brothers to presumably decide to place bombs at the Boston Marathon. One theme that has come up repeatedly, especially with the younger brother Dzhokhar Tsarnaev, is how could he seem so "normal" to his high school and college friends, yet help perpetrate the most publicized act of domestic terrorism since September 2001.
However, research on the self-concept has shown that normal people can possess multiple, context-dependent self-aspects and that these distinct self-identities can be highly variable rather than consistent within a given individual.
The public seems puzzled to hear that Dzhokhar Tsarnaev was described by fellow students as "just like any other college student," as "helpful," and as a "partier who smoked a lot of marijuana." None of these descriptions seem to fit the stereotype of a terrorist who should be an intense loner who is a confrontational ideologue with a big chip on one's shoulder. But even among normal people (and I'm certainly not suggesting that either of the Tsarnaev brothers were "normal" in terms of mental health), having very diverse facets of the self is quite common.
Self-complexity and its implications
For example, it's not unusual to discover that a neighbor who seems to be a warm and caring parent and spouse can be an aggressive cut-throat in a corporate boardroom. In psychology, we examine this within-person variability in a phenomenon called "self-complexity" (McConnell, 2011).
People are greater in self-complexity when they have more self-aspects (e.g., roles, identities, relationships) that are relatively different (e.g., caring as a parent, confrontational at work). Although some people are low in self-complexity (i.e., few identities and quite consistent in their roles), people greater in self-complexity can seem like "completely different people" in different social contexts.
When people are greater in self-complexity, a number of consequences result. First, these people experience less potent emotional reactions in general. In some of our research, we have shown that feedback (positive or negative) stays "localized" within a particular self-identity for people greater in self-complexity (McConnell et al., 2009). Thus, the anger of losing a big client at work might not be revealed by high self-complex people when they come home and see their family. In essence, the differentiation among their selves keeps one world bottled up from another. If one has a terrorist identity, it's easy to imagine that they might not reveal those qualities as a student or with their friends (especially when those qualities might be viewed as undesirable by those audiences).
A second thing we have observed in our research is that people who possess core attributes (what we call "chronic traits") may not exhibit them in contexts where that attribute is not relevant for that particular self-identity (Brown & McConnell, 2009). For example, if "caring" is the most important trait one possesses, our research has shown that they will not show evidence of it in a particular domain where "caring" is not associated with the context (e.g., at work) despite the fact that "caring" might be important in many other domains (e.g., as a spouse, as a parent, with neighbors and friends). In short, for those greater in self-complexity, one's most central quality may be "turned off" in a particular context rather than "always on" and guiding one's actions.
Summary
I'm certainly not arguing that self-complexity explains what would compel fellow citizens to plant bombs to kill innocent people and attempt to kill cops who are trying to protect public safety. However, work on self-complexity suggests that people can act in very inconsistent ways in different contexts, and that such behavior can happen even in "normal" people. Further, living in a culture that emphasizes individualism, we typically exaggerate the degree to which any person has "a true essence." Yet, when we are unaware of this context-dependence of human actions, we end up with lots of questions (e.g., how could a quiet student be caught up in a terrorist plot) that reflect relatively simplistic assumptions about human nature.
References
Brown, C. M., & McConnell, A. R. (2009). When chronic isn’t chronic: The moderating role of active self-aspects. Personality and Social Psychology Bulletin, 35, 3-15.
McConnell, A. R. (2011). The Multiple Self-aspects Framework: Self-concept representation and its implications. Personality and Social Psychology Review, 15, 3-27.
McConnell, A. R., Rydell, R. J., & Brown, C. M. (2009). On the experience of self-relevant feedback: How self-concept organization influences affective responses and self-evaluations. Journal of Experimental Social Psychology, 45, 695-707.
