Rachman focused on a specific cognitive bias called thought-action fusion (TAF), where a person believes that thoughts can directly influence events/action in the real world, increasing the sense of personal responsibility that is common in many people with OCD.

The following study aimed to test the metacognitive explanation of OCD, i.e. the theory that what we think about our thoughts plays a causal role in the development of OCD symptoms. This idea was investigated by inducing thought-action fusion (TAF) and observing the extent to which this affected subsequent intrusive thoughts and associated emotions. It was hypothesized that TAF would increase the frequency of intrusive thoughts and the distress they caused, and this effect would be greater in people who were already prone to obsessional symptoms compared with those who were not.

The study employed an independent samples design whereby participants who were high or low in obsessional symptoms were randomly allocated to either the experimental or control condition. All participants were hooked up to an EEG machine and told that it would detect activity in their hypothalamus if they thought about drinking.

273 university students completed an initial test of OCD symptoms. 32 students in the top and bottom quartiles were recruited, giving a total sample of 64 students.

Firstly, participants completed a questionnaire to assess their current mood to ensure that both groups entered the study in a similar frame of mind. Next, they were given the standardized instruction about the EEG machine which supposedly would detect any thoughts about drinking or water. The participants put on the EEG electrode cap, which was attached to the EEG machine, although this was in fact switched off. Participants then put on headphones and watched three short video clips about drinking water while the researcher supposedly checked to ensure that the EEG was registering hypothalamus activity as expected. They were then told the machine was working well and to sit and think for 4 minutes.

Those in the experimental group were told that the machine was set up so that if they thought about water or drinking, a loud aversive noise might sound. Those in the control group were told the same information and that they may hear the loud noise but this was not in any way linked to their thoughts about water or drinking. After 4 minutes all participants completed self-report measures to assess the dependent variables: the self-reported number of intrusions about drinking, the level of discomfort from these intrusions, the amount of effort to control thoughts about drinking, and the amount of time spent thinking such thoughts.

For participants who scored high on obsessive thoughts, the frequency of intrusive thoughts was greater in the experimental group (6.5) than the control group (3.3), yet there was little difference in frequency of intrusive thoughts in the low obsessional symptoms group, between the two conditions: 5.75 in the experimental condition compared with 5.0 in the control condition.

Similarly, the high obsession symptoms group experienced more distress when they believed their thoughts controlled the noise than when they believed their thoughts were unconnected to the timing of the noise, yet in the low obsessional group more distress was experienced when they thought they had no control over the noise.

These findings suggest that assessing TAF beliefs and targeting them in therapy could help people with OCD to reduce intrusive thoughts and discomfort, and thereby reduce the need for neutralizing behaviours. 

The researchers used a quasi-experimental method. Participants were allocated to groups based on their level of obsessive thinking. Although an independent variable was manipulated, it is not possible to conclude a true causal relationship.

Neuroscience students were not included as they may have seen through the deception - that is, they would have known that thoughts of drinking could not be detected by an EEG. This enhanced the study's internal validity.

The researchers incorporated a post-study check to ensure that participant believed in the deception regarding the EEG detecting thoughts and causing an aversive noise. This suggested that the participants did believe what they were told by the researchers and there was no difference in the level of belief between the different condition.  This suggests that any differences in outcomes could not be attributed to demand characteristics.

The researchers note that the people in the high obsessional symptoms group were not actually diagnosed with OCD, and it would therefore be interesting to replicate the study with a clinical sample. This said, the researchers also state that they did not know whether anyone in the study had been diagnosed with OCD or other psychological difficulties in the past.

Although the study tells us about how thought-action fusion may link to intrusive thoughts and negative emotions, it does not help to explain the urge to neutralize through compulsive behaviours, which is a hallmark feature of OCD.