درباره تبدیل شدن به خنثی: اثرات بازنگری خنثی کننده تجربی

Behaviour Research and Therapy 34 (1996) 889–898 found that writing out a negative thought produced anxiety and an urge to neutralize the thought, that instructing participants to neutralize the thought reduced anxiety/neutralization urge in the short run (i.e. within 2 min), but that in the control group 20 min without instruction was attended by the same reduction in anxiety/urge to neutralize (“natural decay”). The observations were made with pariticipants who scored high on “thought action fusion” and the experiment was set up as exerimental model of obsessions. We repeated the study with participants that were not selected on thought action fusion. All the findings reported by Behaviour Research and Therapy 34 (1996) 889–898 were replicated. Correlational analysis indicated that the strength of the effect was not related to scores on scales measuring “thought action fusion”. Behaviour Research and Therapy 34 (1996) 889–898 did not assess whether non-neutralizing was followed by immediate reductions in distress. We did assess this and found that the larger part of the immediate reduction of distress after neutralization also occurs when no neutralization instruction is given. The effects of neutralization instructions in the present type of experiment are considerably less powerful than suggested earlier.

In order to better understand why Obsessive–Compulsive Disease (OCD) patients are so upset about negative intrusions and why they tend to try and neutralize them, Rachman and co-workers related intrusions and neutralization to “thought action fusion” (TAF) (Rachman, 1993). People with high TAF are “inclined to feel that their unacceptable thoughts may increase the probability of an adverse event occurring (TAF Likelihood) and/or that such thoughts are morally equivalent to carrying out the corresponding unacceptable action (TAF Morality)” (Rachman, Shafran, Mitchell, Trant, & Teachman, 1996, p. 890). To the degree that TAF is present, one would feel responsible for unpleasant thoughts about mishappenings. OCD patients score especially high on TAF and this would help to explain why, relative to non-OCD participants, the former are so upset about having negative intrusions (Rachman & DeSilva, 1978 and Salkovskis & Harrison, 1984) and why they are so much inclined to try and neutralize negative intrusions. Neutralization of negative thoughts is held to be functionally equivalent to carrying out overt compulsions. Both are held to reduce distress in the short run. But in the longer run not overtly undoing and not covertly neutralizing may both just as well be attended by anxiety reduction. As for compulsions, there is indeed data showing that anxiety goes down immediately after overtly undoing and that it goes down in the longer run without overtly undoing ( Rachman & Hodgson, 1980). But what about covert neutralizing? Does immediate emotional relief attend neutralizing likewise, while non-neutralizing is, in a long(er) run, attended by a comparable decay? Rachman et al. (1996) set out to answer this question. First a group of 63 high TAF students was recruited. They were asked to write down the sentence “I hope that X is in a car accident” where X related to a friend or relative who was close to the participant. After writing out this sentence, anxiety and urge to neutralize was measured. On a Visual Analog Scale (VAS), anxiety was much higher than right before writing out the sentence (scores went up from around 15 to around 65). These high TAF participants expressed a strong subjective urge to neutralize. Then the group was split. Half were invited to neutralize, e.g. by altering the sentence by adding “not” or destroying the paper on which the sentence was written. After 2 min of neutralizing, subjective anxiety and urge to neutralize was measured again in this subgroup. Scores dropped dramatically to around 25. The other group was asked to do nothing and after 20 min they too were asked to indicate their level of anxiety and urge to neutralize. Both dropped in the 20 min period; anxiety dropped to the same degree as 2 min neutralizing, while the drop in neutralizing urge was somewhat less in the non-neutralizing group. Rachman et al. conclude that, in high TAF participants, formulating a negative thought produces distress and an urge to neutralize, that a neutralizing ritual immediately reduces distress and urge to neutralize but that non-neutralizing will allow for a “natural decay” of distress and neutralizing urge. The Rachman et al. (1996) study was an attempt to experimentally reproduce crucial elements of obsessive neutralizing. Because of the theoretical importance of this endeavor and its intriguing outcome, a first goal of the present experiment was to see whether the main findings could be replicated. In order to “maximize the effects of an experimental manipulation of neutralizing” (p. 90) Rachman et al. (1996) only tested high TAF participants. It remains unclear whether the observed effects of writing out and neutralizing upsetting thoughts are specific to this group or whether the effects occur regardless of individual differences. Therefore, a second goal of the present experiment was to document whether or not the pattern found in the Rachman et al. data is specific to persons with “high TAF”. Rachman et al. (1996) maintained that ritually neutralizing, just like washing rituals in the case of compulsive cleaners (Rachman & Hodgson, 1980), reduces distress in the short run but that in the longer run “natural decay” (Rachman et al., p. 896) also allows for reduction in distress. The hypothesis boils down to group differences in time course: a quick reduction after neutralization and slow reduction in the case of non-neutralizing. When time-course differences are to be tested, the variable of interest should be measured in the experimental and control groups at the same moments in time. This was not done in the Rachman et al. study. Effects in the neutralization group were tested after 2 min but in the control group, reduction in distress was only established after 20 min. It remains possible that this “natural decay” had already achieved (parts of) its effects within 2 min. The Rachman et al. data do not show that neutralizing has effects over and above “natural decay”, and if it does, how the magnitude of the immediate effects compare. The third goal was therefore to repeat the experiment, but now to test both groups after 2 and 20 min.

The pattern in the present data is graphically represented in Fig. 2. From Fig. 2 it was apparent that after 20 min in both groups anxiety was back to baseline while the urge to neutralize was considerably reduced. Therefore, only the data from the first 20 min were statistically analyzed. Full-size image (8 K) Fig. 2. Anxiety and urge to neutralize after the provocation, after the neutralization instruction (group 1) and in the absence of a neutralization instruction (group 2). Figure options 3.1. Replications 3.1.1. Does anxiety increase and does an urge to neutralize occur after writing out the negative thought? Fig. 2 shows that writing out the negative thought resulted in sharp increases in anxiety in both groups. A two-way ANOVA on anxiety comparing Time (before vs after writing) and Group (group 1 vs group 2) showed a clear effect of Time (F=210.2; df=1,77; p<0.001) but neither an effect of Group (F=1.2; df=1,77; p=0.27) nor an interaction-effect of Time×Group (F=2.8; df=1,77; p=0.1). Fig. 2 shows that both groups expressed an urge to neutralize of about 60 on the 100 Mm.VAS. The intensity of the induced urge did not differ between groups (t=0.94; n=79; p=0.35) 3.1.2. Is neutralizing attended by immediate reduction in anxiety and urge to neutralize? As is clear from Fig. 2, anxiety in group 1 declined after 2 min of neutralizing. This reduction was significant (t=9.3; n=39; p<0.001). The same pattern was observed for urge to neutralize (t=6.7; n=39; p<0.001). 3.1.3. Is not neutralizing, in the long(er) run, attended by reductions of anxiety and urge to neutralize (“natural decay”)? Fig. 2 illustrates that after 20 min of not neutralizing in group 2 (VAS 3), anxiety was lower than immediately after writing out the sentence (VAS 1). The reduction was significant (t=10.8; n=40; p<0.001). The urge to neutralize dropped likewise (t=9.8; n=40; p<0.001). 3.1.4. Are reductions in anxiety and urge to neutralize comparable after 2 min of neutralizing vs 20 min of not neutralizing? A t-test was carried out on the difference between the 2-min reduction of anxiety in group 1 (VAS 1 minus VAS 2) vs the 20-min reduction in group 2 (VAS 1 minus 3). The reductions were comparable (t=0.3; n=79; p=0.8). Likewise, the drop in urge to neutralize after a short time neutralizing in group 1 was not larger than after 20 min of non-neutralizing (t=−0.7; n=79; p=0.49). Thus, the main findings from the original Rachman et al. (1996) study were replicated with non-TAF selected participants. Writing out the negative thought resulted in an increase of anxiety and the emergence of an urge to neutralize (3.1.1, 3.1.2, 3.1.3 and 3.1.4), actual neutralizing was attended by an immediate reduction of anxiety and neutralization urge (Section 3.1.2), but in the longer run the non-neutralization group showed the same effects (Section 3.1.3). Two minutes of neutralizing and 20 min of “spontaneous decay” had the same effects on anxiety. The only difference between the Rachman et al. data and the present one is that in the original study neutralization urge dropped substantially in the 20 min non-neutralizing period, but it remained significantly higher than after 2 min actual ritualizing. In the present study effects of “natural decay” and neutralizing were not significantly different (Section 3.1.4). 3.2. Are the observed results related to TAF scores? The Rachman et al. (1996) pattern may be replicable in non-TAF selected participants, but it remains possible that the effects are more marked when TAF status is taken into account. From TAF theory one should predict that TAF is related to (a) greater increases in anxiety due to writing out the thought, (b) greater drops in anxiety/urge after neutralizing the thought and (c) less spontaneous decay in the non-neutralizing group. Over both experimental groups, there was no correlation between TAF scores (neither total TAF scores nor sub scores) and increase in anxiety from baseline to immediately after writing out the thought (r<0.15; p>0.19). Within group 1, there was no association between the reduction in anxiety from VAS 1 to VAS 2 (correlations of anxiety reduction with all TAF subscales and with TAF total score (r<0.2; p>0.16). Likewise, the drop in urge to neutralize from VAS 1 to VAS 2 in group 1 correlated neither with any of the TAF subscales nor with TAF total score (all correlations: r<0.12; p>0.49). The predicted negative correlation between TAF (sub) scores and effects of “natural decay” did not materialize: all r values<0.1; p>0.5. Thus, while the pattern observed in the Rachman et al. (1996) data was successfully replicated in a non-TAF selected sample (see questions 3.1.1 to 3.1.4), the present data indicate that TAF did not potentiate the effects of writing out the thought or neutralizing it afterwards. 3.3. Is the immediate reduction in anxiety and urge to neutralize stronger after 2 min neutralizing relative to 2 min non-neutralizing/natural decay? Fig. 2 shows that in group 2, within 2 min and without any neutralizing instructions, anxiety and urge to neutralize dropped substantially. The scores immediately after writing out the sentence (VAS 1) were somewhat higher in group 1. Therefore the crucial question was answered by conducting one-way ANOVAs comparing both groups on anxiety/urge 2 min after formulating the thought (VAS 2) which the post-provocation scores (VAS 1) serving as covariate. The group effect was significant for anxiety (F=4.2; df=76; p=0.05) as well as for urge to neutralize (F=6.1; df=76; p=0.02) indicating that neutralizing had an effect over and above the effect that was observed without any instruction. To gain an impression on the magnitude of the 2-min-effects of “natural decay” vs neutralizing, effect sizes were calculated for both groups. Neutralization in group 1 was attended by a d′=1.49 drop in anxiety. Meanwhile some two-thirds of the effect was also observed in group 2: d′=1.1. While groups differed in drops in urge, due to the smaller standard deviation in the non-neutralizing group the effect sizes were virtually identical in both groups. (Neutralizing related drop in urge d′=1.07; effects of 2 min non neutralizing: d′=1.08). A comprehensive comparison of anxiety scores from the Rachman et al. (1996) study and from the present study is given in Table 1. Table 1. Mean VAS anxiety scores from the Rachman et al. (1996) study and from the present study Before provocation Immediately after provocation 2 min after provocation 20 min after provocation Rachman et al. (1996) Group 1 16 65 (Neutralization) 24 13 Group 2 12 70 (No neutralization) — 15 Present study Group 1 12 62 (Neutralization) 24 13 Group 2 13 52 (No neutralization) 29 13