[CogSci Summaries home | UP | email]

Greene, J.D., Nystrom, L.E., Engell, A.D., Darley, J.M., & Cohen, J.D. (2004). The neural bases of cognitive conflict and control in moral judgment. Neuron, 44, 389-400.

author = {Greene, Joshua D. and Nystrom, Leigh E. and Engell,
Andrew D. and Darley, John M. and Cohen, Jonathan D.},
title = {The Neural Bases of Cognitive Conflict and Control in Moral Judgment},
journal = {Neuron},
year = {2004},
volume = {44},
pages = {389-400},
month = {October}

Author of the summary: Michelle Duguay, 2012, meduguay@connect.carleton.ca

Cite this paper for:

The actual paper can be found at: http://www.wjh.harvard.edu/~jgreene/GreeneWJH/Greene-etal-Neuron04.pdf

There are two main theories in moral psychology: the rationalist theory and the emotivist theory. Rationalist theory states that “higher cognition” and reasoning processes determine moral judgements. Emotivist theory states that emotions determine moral judgments.

There are two types of moral judgment: personal and impersonal.
A personal moral violation must fulfill three criteria:

  1. “The violation must be likely to cause serious bodily harm
  2. This harm must befall a particular person or set of persons
  3. The harm must not result from the deflection of an existing threat onto a different party.” [p.389]
Failure to fulfill all three criteria is considered impersonal. [p.389]
An example of an impersonal moral dilemma is the trolley dilemma. An example of a personal moral dilemma is the footbridge dilemma. (see summary author’s notes)

The theory proposed in this paper suggests creating a hybrid of both theories where impersonal moral judgments are regulated by “cognitive” processes and personal moral judgments are regulated by social-emotional processes. [p.389]

This theory could have been the result of evolution. It is presumed that our ancestors relied solely on emotions regarding moral judgments, thus it is believed that they had a “domain-specific, social emotional” system. [p.389] It is believed that, through evolution, humans developed a “domain-general capacity for sophisticated abstract reasoning.” [p.390] Thus, since both systems still exist; they are both used for moral decision making and sometimes come into conflict with each other.

Previous Findings
A suggested interpretation of the results conclude that our primitive emotion-based decision making system causes us to react to personal moral dilemma negatively, thus personal moral dilemmas are initially viewed as inappropriate. In order to overcome this social-emotional response in order to preserve ‘the greater good,’ abstract reasoning and cognitive control systems are activated, enabling a personal moral dilemma to be judged appropriate. [p.390]

  1. Hypothesis for personal and impersonal moral dilemmas was supported.
    a) Brain areas associated with emotion and social cognition were highly active when participants were engaged in personal moral dilemmas. Brain areas associated with emotion and social cognition that were activated during personal moral dilemmas include the “medial prefrontal cortex, the posterior cingulate/ precuneus, and the inferior parietal lobe. [p.391]
    b) Brain areas associated with cognition (abstract reasoning and problem solving) were highly active during impersonal moral dilemmas. Brain areas associated cognitive processes that were activated during impersonal moral dilemmas include the “right dorsolateral prefrontal cortex (DLPFC) and bilateral inferior parietal lobe.” [p.391]
  2. Reaction time (RT) for moral dilemmas judged appropriate or inappropriate.
    a) RT longer when personal moral dilemmas was judged appropriate vs. when judged inappropriate.
    b) These results were not found for impersonal moral dilemmas. [p.390]

Experiments Conducted for this Paper
There are three components outlined in this paper. A replication the previous experiment for the hypothesis regarding impersonal versus personal moral dilemmas was repeated again with a larger sample size. [p.391] Two new hypotheses were created and tested, labelled Analysis 1 and Analysis 2.

General Hypothesis
Conflict between our initial emotional response and our activation of brain areas associated with abstract reasoning and cognitive control will result in an increased RT when faced with a personal moral dilemma. Since the anterior cingulate cortex (ACC) is a brain area linked to cognitive control and the dorsolateral prefrontal cortex (DLPFC) is a brain area linked to abstract reasoning processes, increased activation in these areas during a personal moral assessment should be observed when a participant has a high RT than when a participant has a low RT. [p.390]

Analysis 1
An investigation of personal moral dilemmas is explored regarding differences between difficult and easy personal moral dilemmas. An example of a difficult moral dilemma is the crying baby moral dilemma and an example of an easy moral dilemma is the infanticide dilemma. (see summary author’s notes).

The crying baby dilemma is presumed to be difficult since it elicits a negative, emotion driven response that competes with the cognitive response system. The infanticide dilemma does not involve conflict between competing decision-making systems since the costs far outweigh the benefits, thus, the response to the decision made will lack a cognitive aspect and be driven solely by social-emotional responses. Easy personal moral dilemmas should have a low RT, which means participants should be able to respond quickly to these dilemmas.

It is predicted that the brain areas associated with cognitive control and abstract reasoning, the ACC and DLPFC, should display lower activity levels during low RT, easy personal moral dilemmas than compared to high RT, difficult personal moral dilemmas. [p.391]

Analysis 2
An investigation of response time is explored regarding the judgment of difficult personal moral dilemmas being appropriate versus being deemed inappropriate. A judgment is labelled utilitarian if it entails the judgment of a personal moral dilemma being appropriate and non-utilitarian if it is judged inappropriate. [392]

Since cognitive control is not required for easy personal moral dilemmas, it is presumed that only the ACC should be activated during these dilemmas, not the DLPFC. This analysis wants to determine that the ACC does indeed correlate with abstract reasoning and the DLPFC does indeed correlate with cognitive control. Since it is presumed that both easy and difficult personal moral dilemmas require the activation of abstract reasoning, but only difficult personal moral dilemmas involve a cognitive aspect, only dilemmas such as the crying baby were investigating during analysis 2 to control for the activation of the ACC . [p.396]

It is hypothesized that difficult personal moral violations that were judged appropriate should have a higher RT and more activity in the DLPFC than those that were judged inappropriate. [p.390]

Data was collected from 41 participants. They were put in an fMRI machine and were asked to make judgements of moral dilemmas. [p.398]

Results and Discussion
The results for the replication experiment yielded the same results from the previous paper. (see previous findings section). The larger sample size for this replication included 41 participants. [p.391] As well, activation in the amygdala was observer for personal moral dilemmas, but not in impersonal moral dilemmas. [p.391]

Analysis 1
The results showed increased activity in the anterior DLPFC, ACC, and inferior parietal lobes during difficult personal moral dilemmas compared with activation during easy personal moral dilemmas. There was also activation in the posterior cingulated cortex [p.392]
Mean RT for high-RT trials was 8.38s and mean RT for low-RT trials was 2.83s [p.392]
Thus, the prediction that ACC activity would be increased during difficult personal moral dilemmas was confirmed. [393]
Also confirmed that difficult personal moral dilemmas result in an increase of activity in the DLPFC, which indicated that abstract reasoning and cognitive control compete with social-emotional processes during these types of dilemmas. [p.393]

Analysis 2
The results yielded an “increased activity for utilitarian, as compared to nonutilitarian, moral judgement bilaterally in the DLPFC, right inferior parietal lobe... [and] in the more anterior region of the posterior cingulate.” [p.392]
Thus, a subset of the DLPFC was found to correlate specifically with cognitive control as opposed to abstract reasoning, which suggests that “the neural mechanisms subserving these aspects of cognitive processing may be at least partially dissociable.” [p.396]

Broader Implications
In sum, the general theory that was presented in this paper was confirmed.
“Our present finding that increased “cognitive” activity in the DLPFC predicts utilitarian moral judgment behaviour suggesting that cognitive control processes can override these emotional responses, favouring personal moral violations when the benefits sufficiently outweigh the costs. Thus, both emotional and “cognitive” processes appear to be crucial in producing the patterns of neural activity and behaviour observed in these experiments,” [p.397]
The results of this paper include an integrative third theory for moral psychology to the classical rationalist and emotivist views.

Summary author's notes:

Back to the Cognitive Science Summaries homepage
Cognitive Science Summaries Webmaster:
Jim Davies (jim@jimdavies.org)