Are we in power? The functional neuroanatomy of intentional action
In his talk at the LMU Munich in April 2013, Marcel Brass gives an account of the functional neuroanatomy of intentional control of action. Since he found previous neuroanatomical evidence to be contradictory, he developed a new model decomposing intentional action into what, when and whether components (Brass & Haggard, 2008). Aside from his model, he talks about preconscious determinants of intentional action and discusses the role of free will in it.
The WWW model of intentional action
Marcel Brass starts his talk with showing a video of a patient suffering from the alien hand syndrome. The patient describes that someone was pulling at her hair. After some moments, she realized that it was her own left hand doing it. In general, such involuntary movements don’t seem to be random but rather reactionary to the environment. Among others, Assal et al (2007) have shown that involuntary movements include an activation of the contralateral primary motor cortex, while voluntary movements activate a distributed network implicating not only the contralateral right primary motor cortex and premotor cortex but also the left inferior frontal gyrus and especially the supplementary motor area (SMA) as well as the pre-supplementary motor area (pre-SMA).
In order to map intentional control to more specific brain regions, it is necessary to decompose it in its components. The initiation component is responsible for the timing of the action and the decision component specifies which decision is made. Brass also suggests a third veto component specifying whether an action will actually be performed.
A simple method to investigate the initiation component is the Libet task. Participants sit in front of a computer screen and see a clock hand rotating. They are instructed to press a key at any moment within a given time interval. Afterwards they have to report at which position of the clock hand they decided to press the key, not when they actually pressed the key. The results usually show that the decision to press they key precede the actual pressing by about 200 milliseconds. Interestingly, via EEG recordings it was found that a so-called readiness potential (RP) precedes the time at which the participants reported to have made their decision by about 300ms (e.g. Libet et al, 1983). Brass argues that this RP is related to the SMA and pre-SMA regions. Evidence comes from a microstimulation study by Fried et al (1991). After stimulations of the SMA, participants reported that they felt the “urge” to perform a movement or anticipated that a movement was about to occur. If they were stimulated stronger, they actually performed a movement.
A problem with empirically investigate the selection component is that in order to select from something external stimuli have to be included. The involvement of external stimuli requires a matching on the perceptual and the motor level between the different conditions. To achieve this, Müller et al (2007) designed a new task. It consists of two conditions: externally triggered and intentional. In the first condition, the participants see a cross on the right or left side of the screen and have to press the corresponding arrow key. In the second condition, they freely choose which key to press and a cross appears on the corresponding side of the screen. The fMRI results show that the rostral cingulate zone (RCZ), which is supposed to be activated during conflict resolution, is differently activated in the intentional as compared to the externally triggered stimuli. The preSMA showed equal activity in both conditions. Since conflict only occurs in the internal condition it suggests that the RCZ is involved in decisions between different response alternatives.
Brass discusses it is arguably plausible that there is a third veto component which determines whether the action will be performed or not. Classical inhibition paradigms like the stop-signal or Go/No-Go paradigm rely on external stimuli to stop an ongoing behavior. In order to investigate intentional inhibition of action Brass and Haggard (2007) developed a variant of the Libet task. Participants sit in front of a computer screen and can freely decide when to press a key. Afterwards, they report when they decided to press the key. Up to now, this is the normal Libet situation. However, in about half of the trials, the participants were asked to form the intention to press the key but at the last possible moment inhibit actually pressing the key. Participants could freely choose on which trial they decided to stop the action. Accordingly they had to report when they intended to act. Interestingly, the fMRI results showed activation in the dorsomedial prefrontal cortex in the veto situation. As Brass points out, this region is different from regions, which are activated in situations with different response alternatives, and it is also different from brain regions usually activated in the stop-signal paradigm. This suggests that the intentional inhibition of action can be dissociated from externally triggered inhibition.
Preconscious determinants of intentional action
As already mentioned above, in the Libet experiment, the readiness potential precedes the awareness judgments by roughly 300ms. Libet concluded that the brain decides our conscious intention unconsciously. Many researchers have criticized this. One reason is that it completely relies on the subjective report of the participants’ awareness, which is not necessarily a good measure for intention. In 2008, Soon et al replicated the Libet paradigm and tried to predict intentional action via means of fMRI analyzing the fMRI data with pattern classification. In their variant of the Libet experiment, participants could decide whether they pressed the left or the right key and also when they pressed it. While they are doing this, they see a stream of letters on the screen. Afterwards, they reported which letter they saw on the screen when they decided to press the key. For the analysis, the researchers trained pattern classifiers with their resulting data set. After the training, the classifier is tested against a different data set. Additionally, they were not interested in correctly classifying the motor response so the classifier had to do the classification before the actual motor event. The results reveal, that the decision can be predicted already eight seconds before the action by activity in the lateral frontopolar cortex and about four seconds before the action in the posterior cingulate cortex both with a 60% correctness rate.
In conclusion, this means that simple intentional decisions can be predicted seconds before the participants become aware of their own decision but the prediction rate is relatively low. Also, the brain areas that predict the decisions are different from regions involved in explicit decisions.
The influence of beliefs in free will on intentional action
Some neuroscientists claim that they solved the free will question: it simply doesn’t exist. Others argue that the problem is a philosophical one and can never be answered empirically. As Brass points out, the really interesting question is not whether free will actually exists or not but whether we believe it exists. Does a belief in the non-existence of free will change our intentional actions? Vohs and Schooler (2008) addressed this question by developing a social experiment. Participants were assigned to three groups. The first had to read pro-free will statements, the second pro-determinism statements and the third neutral statements. Afterwards, they had to solve problems and got paid for each problem they solved. While they were doing this, the experimenter excused himself from the room and trusted the participant with paying himself out on his own and afterwards shredding their answer sheet. The results show that the pro-determinism group got paid significantly more than the free will group, the neutral group and experimenter-scored control groups.
But will this high-level manipulation also be reflected in low-level intentional action? Rigoni, Satori & Brass (2011) did a study in the Libet paradigm in which participants had to read either no-free will texts or unrelated texts. The fMRI results reveal that a few hundred milliseconds before participants reported awareness of their decision, there is a difference in the readiness potential between the groups. In conclusion, this suggests that high-level belief manipulation can in fact influence basic neurophysiological potentials.
In summary, we are able to decompose intentional action in different components that are neurobiologically implemented in different regions of the medial prefrontal cortex. It is further possible to predict decisions from brain activity even seconds before participants become aware of their decision. Additionally, high-level beliefs about free will can influence low-level intentional action.
Brass’s research begs the question whether the manipulation of high-level beliefs can influence the predictability of intentional action from brain activation. In further studies, it will also be interesting to investigate why disbelief in free will can lead to antisocial behavior (cheating, overpaying) and what its neural correlates are.
Assal, F., Schwartz, S. and Vuilleumier, P. (2007), Moving with or without will: functional neural correlates of alien hand syndrome. Ann Neurol., 62: 301–306. doi: 10.1002/ana.21173
Brass, M., & Haggard, P. (2007). To do or not to do: the neural signature of self-control. The Journal of Neuroscience, 27(34), 9141-9145.
Brass, M., & Haggard, P. (2008). The what, when, whether model of intentional action. The Neuroscientist, 14(4), 319-325.
Fried, I., Katz, A., McCarthy, G., Sass, K. J., Williamson, P., Spencer, S. S., & Spencer, D. D. (1991). Functional organization of human supplementary motor cortex studied by electrical stimulation. The Journal of neuroscience, 11(11), 3656-3666.
Libet, Benjamin; Gleason, Curtis A.; Wright, Elwood W.; Pearl, Dennis K. (1983). “Time of Conscious Intention to Act in Relation to Onset of Cerebral Activity (Readiness-Potential) – The Unconscious Initiation of a Freely Voluntary Act”. Brain 106: 623–642.
Rigoni, D., Kühn, S., Sartori, G., & Brass, M. (2011). Inducing Disbelief in Free Will Alters Brain Correlates of Preconscious Motor Preparation The Brain Minds Whether We Believe in Free Will or Not. Psychological science, 22(5), 613-618.
Soon, C. S., Brass, M., Heinze, H. J., & Haynes, J. D. (2008). Unconscious determinants of free decisions in the human brain. Nature neuroscience, 11(5), 543-545.
Vohs, K. D., & Schooler, J. W. (2008). The Value of Believing in Free Will Encouraging a Belief in Determinism Increases Cheating. Psychological science, 19(1), 49-54.