Regular articleThe impact of individual differences on the neural circuitry underlying sadness
Introduction
Over the last decade, several functional neuroimaging studies have been conducted to investigate the neural substrate underlying various basic emotions. In all these studies, results were reported as average brain activation patterns for the sample of subjects assessed, while individual data were not presented. Importantly, researchers have recently argued that intersubject variability may be of paramount importance for the outcome of functional brain imaging investigations (e.g., Canli et al 2001, Hunton et al 1996, Hasnain et al 1998, Nadeau et al 1998, Steinmetz and Seitz 1991, Xiong et al 2000.
Because of various constraints, samples in functional neuroimaging studies are usually small (often 10 or fewer subjects). Consequently, individual differences in intensity of activation, spatial extent of activation, and location of regional brain activation can have a significant impact on group results. This is especially true when intersubject averaging methods, such as “fixed-effect” analyses, are used (see Friston et al., 1999). In our view, small sample sizes and individual differences may explain, at least partially, why results from functional neuroimaging studies are often divergent (regarding this issue, see Nadeau et al., 1998).
The literature regarding the neural basis of sadness, one of the basic emotions (Plutchik, 1994), illustrates this point. To date, several functional brain imaging studies have been carried out to delineate the neural circuitry of sadness in healthy subjects (Table 1). Interestingly, although some regions such as the insula and anterior cingulate cortex have been reported somewhat consistently, there are also significant discrepancies in the results of these investigations. Regarding this issue, a metaanalysis recently published by Phan and colleagues (2002) revealed that even the subcallosal cingulate cortex, which is the brain region most consistently reported as being activated in sadness induction studies, was seen activated in only 46% of these studies.
In attempting to understand the inconsistencies in the results of these studies, it may be useful to consider the concept of individual differences. Indeed, Canli et al. (2001) recently reported individual differences in brain reactivity to emotional stimuli. These differences could provide an explanation for the inconsistencies observed in the literature. In addition to individual differences, however, other methodological differences among these studies might also be responsible for these inconsistencies. For instance, results from studies using different brain imaging techniques—functional magnetic resonance imaging (fMRI), positron emission tomography (PET), or single photon emission computed tomography (SPECT)—can vary according to the spatial and temporal resolutions of the brain imaging technologies utilized, as well as their sensitivity.
The inconsistencies could also arise from the use of different populations across studies. In fact, differences have been found between men and women in terms of location and intensity of neural activation during sadness George et al 1996, Pardo et al 1993, Schneider et al 2000. It is unlikely, however, that this factor can account for all the variability observed in the literature, because three of the PET studies described above examined only women George et al 1995, Lane et al 1997a, Liotti et al 2000 and yet, they reported discrepant findings.
It is also important to note that the methods used to induce sadness varied widely across these studies. Whereas some researchers used external stimuli such as film excerpts (Beauregard et al., 1998b), others asked subjects to self-generate sad feelings, usually by recalling sad autobiographical events Pardo et al 1993, Damasio et al 2000, Liotti et al 2000. And in some of these studies, both methods were combined George et al 1995, Lane et al 1997a. Given the evidence that internally and externally generated emotions may recruit slightly different brain regions (Reiman et al., 1997), it is possible that differences in mood-induction procedures are partly responsible for the inconsistencies found in the literature regarding the neural substrates of sadness. Again, however, this explanation alone is insufficient, because different results have been obtained in studies in which the same method was used to induce sadness (e.g., George et al 1995, Lane et al 1997a.
In sum, it is possible that individual differences, more than any other methodological differences, are responsible for the inconsistencies found in the literature regarding the neural substrates of sadness. To examine this hypothesis, we used fMRI in two methodologically identical studies designed to identify the neural correlates of sadness in two groups of healthy female subjects. We predicted that the average brain activation patterns would be significantly different between these two groups. We also predicted that, in both studies, the individual activation maps would be characterized by an important degree of intersubject variability.
Section snippets
Subjects
Ten healthy female volunteers participated in each of the two studies. To ensure sample equivalence, subjects from both studies were selected in the same manner. Inclusion criteria were gender (female), age (20–30 years), language (French speaking), degree of education (university students), manual dominance (right-handed), and having no history of psychiatric or neurological disorder. Mean age of participants was 24.1 (age range: 20–27) in Study 1 and 24.5 (age range: 22–30) in Study 2 (t (18)
Behavioral data
In both studies, subjects reported feelings of sadness during the Sad condition (average rating for Study 1 = 5.33, SD = 1.5, range, 2–7; average rating for Study 2 = 4.75, SD = 0.75, range, 4–6), but not during the Neutral condition (average rating for each study = 0). The difference between the average sadness ratings reported in Study 1 vs Study 2 was not significant (t (18) = 0.65, P > 0.05). While viewing the sad film clips, a few subjects also reported feelings of anger (average rating
Discussion
FMRI was used in two separate studies, and in two different groups of healthy female subjects, to identify the brain regions activated during a transient experience of sadness induced by the viewing of sad film excerpts. Methodological procedures for both studies were identical and subjects were recruited using the same criteria. On average, the level of sadness reported by each group was equivalent. Despite all this, the random-effects analyses yielded different results for each study.
Conclusions
The present results showed important variability in individual brain activation patterns during transient sadness induced by film excerpts. These results suggest that, even in the absence of other methodological differences, different brain structures can be identified as neural correlates of sadness in functional neuroimaging studies using different but comparable groups of subjects. These findings strongly support the view that individual differences may be responsible for the inconsistencies
Acknowledgements
This work was supported by grants from National Sciences and Engineering Research Council of Canada (NSERC), National Alliance for Research on Schizophrenia and Depression (NARSAD), and Département de radiologie, Faculté de médecine, Université de Montréal to M.B., and by a studentship from National Sciences and Engineering Research Council of Canada (NSERC) to F.E. We thank the staff of the Département de radiologie, Centre hospitalier de l’Université de Montréal (CHUM), Hôpital Notre-Dame,
References (34)
- et al.
The functional neuroanatomy of amusement, disgust, and sexual arousal
NeuroImage
(1998) - et al.
A functional anatomy of anticipatory anxiety
NeuroImage
(1999) - et al.
Anger in healthy men: a PET study using script-driven imagery
Biol. Psychiatry
(1999) - et al.
How many subjects constitute a study?
NeuroImage
(1999) - et al.
Gender differences in regional cerebral blood flow during transient self-induced sadness or happiness
Biol. Psychiatry
(1996) - et al.
Regional brain activity during transient self-induced anxiety and anger in healthy adults
Biol. Psychiatry
(1999) - et al.
Neuroanatomical correlates of pleasant and unpleasant emotion
Neuropsychologia
(1997) - et al.
Affect intensity as an individual characteristic: A review
J. Res. Personality
(1987) - et al.
Differential limbic-cortical correlates of sadness and anxiety in healthy subjects: implications for affective disorders
Biol. Psychiatry
(2000) - et al.
Functional neuroanatomy of emotion: a meta-analysis of emotion activation studies in PET and fMRI
NeuroImage
(2002)