Background & Aims
The affective-motivational system is involved in both pain perception and pain avoidance. Avoidance response can either be to inhibit behavior causing pain (positive punishment, PP) or facilitate behavior to obtain pain relief (negative reinforcement, NR). Therefore, pain is part of two types of operant learning. The inclination to engage in operant learning to obtain pain avoidance and pain relief depends on the personal experience of pain and by the neural processes happening during a painful sensation. In operant conditioning, Skinner found three operants: neutral (probability of a behavior to be repeated does not change), reinforcer (increase the probability of a behavior to be repeated) and punisher (decrease the probability of a behavior to be repeated). [1][2] [3]
The aim of this study was to examine alterations in the event related potentials (ERPs) following an auditory cue indicating if the performance of a proceeding task results in PP or NR.
Methods
29 subjects rated a pressure ramp induced by a cuff algometer on a VAS scale (0=no sensation, 10=worst pain, 5=pain threshold (PT)). The pressure remained at PT during a cognitive task (repeat 5-word sentences with background noise to balance the correct/incorrect responses to 50%). The experiment had 2 conditions: negative reinforcement with pain relief (VAS=3) as reward (NRr) and unaltered PT as punishment (NRp); and positive punishment with increased pain (VAS=7) as punishment (PPp) and unaltered PT as reward (PPr). Distinct auditory feedback stimuli for correct/incorrect responses were given before the punishment/reward. Subjects did 120 trials in 3 sessions with 4 randomized blocks (subjects were informed about the condition). EEG data from 64 channels were recorded and pre-processed with EEGlab. ERPs were averaged over all trials and epoched from -500 to +1000 ms post-auditory feedback. Complexes P1N1, P2N2 and P3N3 were identified in Cz electrode and analyzed using an RM-ANOVA.
Results
In the Negative Reinforcement condition (NR), the peak-to-peak amplitudes for incorrect (p) was higher than correct (r) in P1N1 (p<0.001; mean: NRp=18.99±8.56 µV, NRr=15.79±6.62 µV) and P2N2 complexes (p<0.01; NRp=20.40±9.24 µV, NRr=12.23±4.21 µV). In the P3N3complex no differences were found (NRp=14.98±6.66 µV, NRr=14.33±5.76 µV). In the Positive Punishment condition (PP), the peak-to-peak amplitude for incorrect (p) was higher than correct (r) in all the complexes: P1N1 (p<0.01; PPp=18.07±6.19 µV, PPr=14.31±6.69 µV), P2N2 (p<0.001; PPp=17.96±5.05 µV, PPr=10.86±5.02 µV), P3N3 (p<0.05; PPp=16.56±6.89 µV, PPr=12.80±6.06 µV). Following a correct performance, NR had increased amplitude compared to PP in the P1N1 complex (p<0.05). The total number of words repeated correctly in the NRp (39.69±10.87) condition was lower than the total number repeated in the PPp (60.89±17.47) with p<0.001.
Conclusions
In all the complexes of the ERPs extracted, the peak-to-peak amplitudes were higher after an incorrect answer (NRp>NRr, and PPp>PPr). The only exception is the P3N3 complex in the NR condition.
Comparing the conditions with a correct answer (NRr and PPr), the peak-to-peak amplitudes of the P1N1 complexes resulted larger in NRr, meaning that the response was bigger when subjects expected a change in the painful stimulation, in this case when there was a pain relief.
From the behavioral data, it was possible to observe that subjects performed better in trying to repeat more words in the PP condition in order not to have higher pain stimulus.
References
[1]F. Porreca and E. Navratilova, “Reward, motivation, and emotion of pain and its relief,” Pain, vol. 158, no. 4, pp. S43–S49, 2017, doi: 10.1097/j.pain.0000000000000798.
[2]T. J. K. Ben Akpan, “Classical and Operant Conditioning—Ivan Pavlov; Burrhus Skinner,” in Science Education in Theory and Practice An Introductory Guide to Learning Theory, 2021, pp. 71–84.
[3]B. F. Skinner, “SCIENCE AND HUMAN,” 2014.
Presenting Author
Carolina Ceruti
Poster Authors
Topics
- Pain Imaging