Emotions and Reidentification

    The basic condition for reidentification, defined in proposition (67), makes no reference to emotional states that are observed while events confirm or disconfirm people's situational identities. Such a formulation is useful because in many instances emotional states are unknown, poorly appraised, or purposely discounted. In other cases, though, such a formulation is too simplistic.

[W]hen we are present at the events that stimulate reidentifications or when we observe a person narrating his or her own behavior, we have access to the expressive signaling system that our species has evolved, and we obtain information about what emotions the person feels as a result of the events. ... Reassessment of the person's character must take account of the emotion displayed because the expressive behavior may cue us that the other feels disconfirmed by events rather than confirmed. (Heise, 1989, p. 14)

This section delineates two approaches to incorporating information on emotions into reidentification processes. The first presumes that the target individual is in a particular mood while experiencing an event, and an observer tries to factor that mood into an inference about the identity that the target individual must have. This formulation was introduced by Heise (1989). In the second approach, an observer tries to infer the identity of the target individual from seeing the emotion that a particular event triggers in the target individual. This formulation is presented in the 2000 revision of this web document for the first time.

Inferences From Mood

    A mood is an emotional predilection being maintained as an aspect of one's identity in a particular situation. The emotion defining the mood amalgamates with one's role identity, generating the fundamental self to be maintained in the situation. Equation  (62c) provides the relevant formulation of the amalgamation process for present purposes:

     (75)

q is the amalgamate EPA profile, e is the EPA profile for the emotion defining the individual's mood, and r is the EPA profile for the individual's role identity. Vector d provides equation constants, and 3x3 matricies E and R consist of coefficients for weighting the first-order contributions of emotion and role identity to the amalgamate. I_Ee is a 3x3 diagonal matrix with the emotion evaluation in the diagonal cells, I_Ee is a 3x3 diagonal matrix with the emotion potency in the diagonal cells, and I_Ee is a 3x3 diagonal matrix with the emotion activity in the diagonal cells, and the Q matricies consist of coefficients for weighting the second-order interactions of emotion and identity EPAs in generating the amalgamate.

    The vector of EPA values for the amalgamate can be substituted in equation (8)  for the fundamental profile of the actor in an event. 

     (76)

or 

     (77)

In this application, r' = [_{e p a} ], the fundamental EPA profile of the actor's identity, and can be expressed as a function of the original f vector in equation (8)

     (78)

by means of the following definitions:

     (79)

and

     (80)

    Equation (12), which interprets unlikelihood as a sum of deflections, can be rewritten using as follows:

     (81)

Substituting the expression for from (86) gives

[Section above revised and shortened October, 2000.]

(90)

or

(91)

where 0 is a vector of zeros whose order matches the order of t. Multiplying out gives

(92)

or

(93)

    At this point we use equation (15) to express t in terms of transients existing before the event:

(94)

or

(95)

This reduces to

(96)

Since the mood is treated as a given, phi is constant during the reidentification, and the last term may be absorbed into the equation constant. Doing this and rearranging terms gives

(97)

which has exactly the structure of (18), the generalized equation for unlikeliness. Thus the solution for the optimal actor is given by (47) with H in equation (29) redefined as

(98)

and h in equation (30) as

(99)

    Heise (1989) pointed out an interesting property of the solution: the predicted evaluation of an actor is positive if the actor's emotion is consistent with the impact of the actor's behavior (e.g., feeling ashamed about an act with bad consequences or joyous about an act with good consequences). The source of this effect is not obvious in the equations above, but I will return to the issue when presenting an illustrative analysis.

Inferences From Emotion

Instead of interpreting q in equation (62c) as a fundamental self created from an amalgamation of mood and role, we can interpret q as an individual's transient state of being formed from the amalgamation of an emotion and role to match the transient state of being created by an event. 

Suppose that the individual is the actor in the event. Then q' = [A_e A_p A_a]'. Additionally from equation (15), [A_e A_p A_a]' = t'M_A, where M is the array of impression-formation coefficients, and M_A refers to the portion that is used to predict impressions of an event's actor. Then q' = t'M_A, and from (62c)

     (99a)

Interpret r as the individual's fundamental EPA profile based on the individual's role identity. Then the question of what role identity the individual must have in order to feel the given emotion as a result of the given event is answered by solving for r.

     (99b)

The event is represented in t, which is made up of the pre-event transient EPAs for actor, behavior, and object, along with interaction terms involving these transients.