In pursuing our studies of the affective states generated by visual inputs, we concentrate in this work on the sentiment of hate. Like romantic and maternal love which we reported on previously (Bartels and Zeki 2001, 2004) , , hate is a complex biological sentiment which throughout history has impelled individuals to heroic as well as evil deeds. Unlike romantic love, it need not be directed against an individual; it may instead assume many varieties, being directed against an individual, a society, or an ethnic group. In this study, we were interested to explore the neural correlates of hate directed against an individual. There are varieties even within such a confine. The hatred may be directed against a public figure or a personally known individual, for a variety of reasons.
This difference in the extent of deactivated cortex, compared to the deactivated cortex in the context of romantic love, may seem surprising, since hate too can be an all consuming passion. But whereas in romantic love, the lover is more likely to be less critical and judgmental regarding the loved person, it is more likely that in the context of hate the hater may want to exercise judgment in calculating moves to harm, injure or otherwise extract revenge.
In summary, our results show that there is a unique pattern of activity in the brain in the context of hate. This pattern, while being distinct from that obtained in the context of romantic love, nevertheless shares two areas with the latter, namely the putamen and the insula. This linkage may account for why love and hate are so closely linked to each other in life.
A questionnaire was devised to assess each subject’s feelings about the hated person. The questionnaire was based on Sternberg’s  triangular theory of the structure of hate. Sternberg proposes the following three elements of hate:-
A/ Negation of Intimacy (Distancing) in Hate: Repulsion and Disgust
B/ Passion in Hate: Anger-Fear
C/ Decision-Commitment in Hate: Devaluation-Diminution Through Contempt
For each of these components A, B and C we construct three negative statements (1/2/3) and one positive statement (4) as below:-
A1/ I don’t want X anywhere near me.
A2/ The world would be a better place if X had never existed.
A3/ Any time spent with X is a waste of time.
A4/ I would like to interact with X.
B1/ I cannot control my hatred for X.
B2/ I would like to do something to hurt X.
B3/ I have violent thoughts about X.
B4/ I have kind thoughts for X.
C1/ X is scum.
C2/ X is a low class type of person.
C3/ X does not deserve any consideration or compassion.
C4/ X is a very nice person.
There are twelve statements in all. In each statement “X” represents the hated person. For each statement the subject must pick one of the following seven evaluations;
“Strongly agree” “Agree” “Mildly agree” “Don’t know” “Mildly disagree” “Disagree” “Strongly disagree”
The negative statements (1/2/3 for each component) are scored from zero for “Strongly disagree” to six for “Strongly agree”. The positive statements (4 for each component) are scored in the opposite sense. Therefore the questionnaire can result in a total “hate” score ranging from 0 (minimum hate) to 72 (maximum hate).
The questionnaire was completed by each subject once during the laboratory visit and once during the scanning visit. On each occasion the questions were presented in a different randomised order.
The hate questionnaire scores had a possible range of 0 to 72. Subjects scores during the laboratory visit ranged from 36 to 66. Scores obtained directly after scanning (which were used in the second level analysis) ranged from 34 to 72. There were substantial differences (both positive and negative) between some of the scores obtained during the lab visit and those obtained directly after the scan, the largest difference being a drop of 16 in the second visit score.