Common pathways in mental imagery and pain perception: an fMRI study of a subject with an amputated arm.

The present paper reviews data from two previous studies in our laboratory, as well as some additional new data, on the neuronal representation of movement and pain imagery in a subject with an amputated right arm. The subject imagined painful and non-painful finger movements in the amputated stump while being in a MRI scanner, acquiring EPI-images for fMRI analysis. In Study I (Ersland et al., 1996) the Subject alternated tapping with his intact left hand fingers and imagining "tapping" with the fingers of his amputated right arm. The results showed increased neuronal activation in the right motor cortex (precentral gyrus) when tapping with the fingers of the left hand, and a corresponding activation in the left motor cortex when imagining tapping with the fingers of the amputated right arm. Finger tappings of the intact left hand fingers also resulted in a larger activated precentral area than imagery "finger tapping" of the amputated right arm fingers. In Study II (Rosen et al., 2001 in press) the same subject imagining painful and pleasurable finger movements, and still positions of the fingers of the amputated arm. The results showed larger activations over the motor cortex for movement imagining versus imagining the hand being in a still position, and larger activations over the sensory cortex when imagining painful experiences. It can therefore be concluded that not only does imagery activate the same motor areas as real finger movements, but also that adding instructions of pain together with imaging moving the fingers intensified the activation compared with adding instructions about non-painful experiences. From these studies, it is clear that areas activated during actual motor execution to a large extent also are activated during mental imagery of the same motor commands. In this respect the present studies add to studies of visual imagery that have shown a similar correspondence in activation between actual object perception and imagery of the same object.

Different brain areas activated during imagery of painful and non-painful 'finger movements' in a subject with an amputated arm.

The purpose of the present study was to investigate differences in brain activation with functional magnetic resonance imaging (fMRI) during imagery of painful and non-painful 'finger movements' and 'hand positioning' in a subject with an amputated arm. The subject was a right-handed man in his mid-thirties who lost his right arm just above the elbow in a car-train accident. MRI was performed with a 1.5 T Siemens Vision Plus scanner. The basic design involved four conditions: imagining 'painful finger movements', 'non-painful finger movements', 'painful hand positioning', 'non-painful hand positioning'. Imagery of finger movements uniquely activated the contralateral primary motor cortex which contains the classic 'hand area'. The lateral part of the anterior cerebellar lobe was also activated during imagery of finger movements. Imagery of pain uniquely activated the somatosensory area, and areas in the left insula and bilaterally in the ventral posterior lateral nucleus of the thalamus. It is suggested that the insula and thalamus may involve neuronal pathways that are critical for mental processing of pain-related experiences, which may relate to a better understanding of the neurobiology of phantom limb pain.

Left frontal activation during a semantic categorization task: an fMRI-study.

In the present study we measured brain activation, with functional magnetic resonance imaging (fMRI) during the execution of a covert semantic categorization task. This involves activation of working memory and internal concept generation. Previous brain imaging studies of covert verbal fluency have shown widespread activation in the frontal and temporal lobes, and anterior cingulate. However, most of these studies have employed simple stimulus tasks with repetition of words e.g., beginning with a certain letter of the alphabet. Moreover, the subject is typically cued (either visually or auditory) every 2-5 second. In the present study we used a "single-cue" instruction at the beginning of each activation period where the subject was instructed to internally generate category specific names related to: "States in the USA," "UK Soccer clubs," and "Male names" during 54s periods. The three activation periods were compared to three baseline periods in which the subject was instructed to imagine mentally "lying on a beach and looking at the sky." Functional MRI was performed with a 1.5T Siemens Vision scanner. Initial TIW 3D FLASH scanning of anatomy was done, and thereafter, serial imaging with 60 BOLD sensitive echo planar whole brain measurements were done during the active and passive tasks. Main activation areas were found in the left middle frontal gyrus, partially overlapping with Brodmann area 9. This is in agreement with previous studies of increased activation in the left frontal lobe, and may indicate a left frontal neural network for accessing the mental lexicon.