What is in a word? No versus Yes differentially engage the lateral orbitofrontal cortex

Abstract

The words "No" and "Yes" are involved in conditioning to prohibit or encourage behavior, respectively. The authors, therefore, hypothesized that these words would be attributed to endogenous valence, activating neuronal circuits involved with valence and emotional control. Functional MRI (fMRI) at 4 Tesla was used to record regional brain activity while participants were exposed to emphatic vocalizations of the words. Results showed that No and Yes were associated with opposite brain-behavior responses; while No was negatively valenced, produced slower response times, and evoked a negative signal in the right lateral orbitofrontal cortex (OFC), Yes was positively valenced, produced faster response times, and evoked a positive signal in a contiguous region of the OFC. Attribution of negative valence to No and trait anger control were associated with increased responsivity of the OFC to No. Inasmuch as sensitivity to the prohibitive command No develops during childhood through interaction with primary caregivers as the first social objects, our findings may implicate the lateral OFC in the neurobiology of emotion regulation and subsequent social development.

Figure 1

(top) A graphic scheme of one of the task runs with 8 word blocks alternating with fixation baseline of the same length. The complete task included 4 of these runs each with a different pseudorandomized order of word conditions. (bottom) BOLD images of the general task activations.

Figure 2

Differential brain and behavioral responses to No and Yes (N = 23). (a) Post-scan ratings of valence (−10 most negative to +10 most positive) to the contrasts No-Up (black bar) and Yes-Ten (white) (non-parametric Wilcoxon, Z = −4.6, p < 0.0001). (b) Reaction-times to the contrasts No-Up (black bar) and Yes-Ten (white) (t = −5.5, p < 0.0001). Note, the y axis starts with 400 ms. (c) percent signal change in the right lateral OFC, BA 47, for -No-Up (Talairach coordinates: 30, 33, −3) and Yes-Ten (coordinates: 33, 30, −9), t = −5.9, p < 0.0001. Error bars at a–c represent ± standard error. (d) Corresponding image.

Figure 3

Correlations between behaviors, corrected for age, and BOLD changes (z-scores) in the lateral OFC in response to No. (a) Scatter plot shows association between the BOLD signal change for -No-Up in the right OFC (x = 42, y = 36, z = 9) and valence No-Up with a linear regression line (r = −0.59, p < 0.001) and with an imbedded corresponding image. The left OFC activation was not significant (b) Correlation between control of anger and BOLD signal change for No-baseline in the right OFC (x = 36, y = 30, z = −3) and a linear regression line (r = 0.77, p < 0.001) with an imbedded corresponding image (Left = Right).