Individual Psychological Differences 79particularly by in”ammation resulting from the production of
proin”ammatory cytokines (Kiecolt-Glaser, McGuire, Robles,
& Glaser, 2002). Individual differences in emotional and cop-
ing responses may account for some of the variation in neuro-
endocrine and immunological changes associated with stress.
Currently, research is aimed at identifying the relationships
among these changes and emotional traits and states, daily
subclinical ”uctuations in mood, bereavement, clinical disor-
ders of major depression and anxiety, and coping strategies.
Negative and Positive Affect
Negative affect is de“ned as general subjective distress and
includes a range of negative mood states, such as depression,
anxiety, and hostility (Watson & Pennebaker, 1989). Cohen
and colleagues demonstrated an association between nega-
tive affect and rates of respiratory infection and clinical colds
following intentional exposure to “ve dif ferent respiratory
viruses (S. Cohen, Tyrrell, & Smith, 1991). A dose-response
relationship was found between rates of respiratory infection/
clinical colds and increased levels of a composite measure of
psychological stress that included negative affect, major
stressful life events, and perceived ability to cope with cur-
rent stressors. In further analyses of these data, negative af-
fect predicted the probability of developing a cold across the
“ve dif ferent upper respiratory infection viruses independent
of negative life events (S. Cohen, Tyrrell, & Smith, 1993).
Furthermore, the higher illness complaints in individuals
high in state negative affect were associated with increased
severity of colds and in”uenza as seen in the amount of
mucus produced (S. Cohen et al., 1995). However, negative
affect was not related to the development of clinical colds
among already infected individuals but rather was associated
with individuals• susceptibility to infection (S. Cohen et al.,
1993; Stone et al., 1992).
In another study, baseline personality variables that are
thought to be characteristic of negative affect (high internal-
izing, neuroticism, and low self-esteem) predicted lower
titers of rubella antibodies 10 weeks postvaccination in sub-
jects who were seronegative prior to vaccination (Morag,
Morag, Reichenberg, Lerer, & Yirmiya, 1999). This relation-
ship was not found in subjects who were seropositive prior to
vaccination.
Dispositional positive affect and the expectation of posi-
tive outcomes, termed optimism, have been less well studied
in relation to immune variables. Davidson and colleagues
(Davidson, Coe, Dolski, & Donzella, 1999) demonstrated
positive relationships between NKCC and greater positive
dispositional mood, de“ned by relative left-sided anterior
brain activation. Greater relative left-sided activation was
associated with higher levels of basal NKCC and with
smaller declines in NKCC from a nonstress baseline to a “nal
exam period that occurred six weeks later.
Although optimism has been related to positive physical
health outcomes in surgery patients (Scheier et al., 1999), its
association with immune function has been inconsistent
among prospective studies of naturalistic stressors. These in-
consistencies might be due to different methodology in de“n-
ing optimism, different periods of follow-up for immune
measures, and differences in the presence and de“nition of
acute and chronic stress. Segerstrom, Taylor, Kemeny, and
Fahey (1998) examined optimism and immune function in
“rst-year law students before entry into the law school pro-
gram and again at midsemester, two months before students•
“rst examination period. Dispositional optimism was not re-
lated to immune measures but to higher situational optimism
(de“ned as positive expectations speci“c to academic perfor-
mance) and was associated with higher NKCC. This associa-
tion was partially mediated by lower levels of perceived
stress. In another study, healthy women were followed for
three months, using daily self-reports of stressful events. In
this case, dispositional optimism was associated with a
greater reduction in NKCC following high stress that lasted
longer than one week compared to less optimistic individuals
(F. Cohen et al., 1999). Thus, optimism may have differential
effects on NKCC, depending on whether situational or dispo-
sitional optimism is measured.Daily Negative and Positive MoodThe relationships between normal daily mood ”uctuations
and immune variables have been evaluated by tracking sub-
jects• naturalistic mood changes and by inducing positive and
negative mood states in the laboratory. In the “rst case, nega-
tive mood over the course of two days was associated with
reduced NKCC, but there was evidence that positive mood
moderated this association (Valdimarsdottir & Bovbjerg,
1997). In the second case, studies of induced mood in the lab-
oratory have shown transient increases in NKCC (Futterman,
Kemeny, Shapiro, & Fahey, 1994; Knapp et al., 1992), but
con”icting outcomes related to the lymphocyte proliferative
response to PHA. Both positive and negative induced mood
conditions were associated with a decreased response to PHA
(Knapp et al., 1992), whereas positive induced mood was as-
sociated with an increased response to PHA and negative in-
duced mood was associated with a decreased response to
PHA (Futterman et al., 1994). The differences in immune
outcomes in these two laboratory-induced mood studies may
be, in part, due to different levels of arousal and physical
activity during the mood induction procedure and the use of