Orienting attention for pain attenuation in patients with chronic low back pain : a study of mechanism using event-related potential

Abstract

Orienting attention between one's internal and external environments involves both top-down control and bottom-up control, particularly when the external-to-internal (E-I) difference increases among patients with chronic pain (specifically chronic lower back pain (CLBP)). The three event-related potential components (ERP), i.e., N1 (100-200 ms), P2 (260-380 ms), and P3 (340-400 ms) can be markers reflecting the attention disengagement, shifting and re-engagement sub-processes of E-I orienting attention, respectively. This thesis aims to investigate how the neural processes underlying E-I orienting attention are modulated by 1) the salience level of the external stimulus, 2) the salience level of the internal representation, and 3) the pain experience of CLBP patients. A total of 19 healthy individuals (9 females) volunteered in the first ERP Study and 15 CLBP patients (12 females) volunteered in the second ERP Study. The participants were to perceive a fleeting (50 ms) external nociceptive stimulus (ES) at the ankle in Study 1 and at the ankle (non-painful site (SNP)) and lower back ("painful" site (SP)) in Study 2. The salience level of the ES was either Low (EL) or High (EH). Next, the participants were required to mentally rehearse a Low (IL) or High (IH) salient internal representation (IR) of the self-generated sub-nociceptive image for 3s. This was followed by assigning a numeric rating scale (NRS) score to indicate the pain intensity of the perceived nociceptive stimulus in Study 1 or perceiving another external stimulus and then comparing it with a self-generated/maintained image in Study 2. Electroencephalography (EEG) signals were captured throughout the process. Among the healthy individuals, a three-way repeated measures ANOVA on the amplitudes of the three ERP components in the first study revealed that the ES × IR× electrode interaction was not significant, but the ES × IR interaction effect was significant in all the three components (F(1,28) = 5.781, p= .016 and F(1,28) = 4.947; p= . 025 for the SP3, and SP3/P2 time window of the N1 component; F(28,504) = 2.204, p< .001 for the P2 component; and F(28,504) = 2.374, p<.001 for the P3 component). Further analysis indicated that the differences between the IH image and the IL image were only significant in the EH condition. Besides, a two-way repeated measures ANOVA on the NRS scores revealed that the ES and IR main effects were significant (F(2,36)= 215.80, p = .001 and F(2,36) = 4.17, p = .012, respectively). Additionally, significant and positive correlations between the attenuation in NRS scores and the P3 component were revealed by Pearson correlation analysis (the r-values were from .517 to .638, p< .050). Among the CLBP patients, a four-way repeated measures ANOVA on the amplitudes of all the three ERP components in the second study revealed that 1) the ES × IR × electrode × stimulation site (SS) four-factor interaction effect was not significant, but ES × IR ×SS three-factor interaction was significant (F(2,28) = 3.678, p <.038) in the N1 component; 2) only a marginally significant ES × IR interaction effect (F(2,28) = 3.129, p= .059) was significant for the P2 component; 3) the ES × IR × electrode × stimulation site (SS) four-factor interaction effect was significant (F(16,224) = 2.484, p <.002) for the P3 component. Further analysis indicated that 1) the N1 amplitudes were more negative-going in the SP condition than that in the SNP condition, 2) the P3 amplitudes were more positively-going for an EH stimulus in the IH condition at the SP site, but not the SNP site. Besides, significant correlations between the attenuation in NRS scores and the amplitude changes of all the three components were revealed by Pearson correlation analysis as well (the r-values were from .541 to .652, p< .050). These findings suggest that the sub-processes underlying E-I orienting attention serve different roles. The disengagement sub-process tends to be stimulus dependent, which is bottom-up in nature. Shifting and reengagement tend to be top-down sub-processes, which involve more cognitive control. These sub-processes may account for the attenuation effects on perceived pain intensity after orienting attention.