Page 31 - Nutrinsight-4

NutrInsight • Satiety: News Insights
5.2 Measuring neural activity
Sensory specific satiety measured at the single neuron level
Understanding brain activity in response to food signals is an important but difficult task in living animals. Early approaches examined the activity of single neurons in the brain of laboratory animals, using implanted electrodes in areas of interest. Typically, an animal’s response to foods of varying sensory characteristics and palatability levels was recorded under conditions of hunger and satiety. One classic example of this approach is the study of sensory specific satiety (SSS) in monkeys.
Sensory specific satiety (SSS) is expressed as a reduced acceptance of a food which occurs after the food has been eaten to satiety, in parallel with a reduction in the pleasantness of its taste, while the pleasantness of other foods is maintained. Studies of the activity of single neurons in the brain of a monkey showed that SSS is not associated with a reduction in the response of neurons that reflect simple gustatory stimulation (nucleus of the solitary tract or frontal opercular or insular gustatory cortices). In the secondary taste cortex, however, decreases in the responsiveness of neurons are relatively specific to the food with which the monkey was fed to satiety. For example while a monkey was fed to satiety with repeated 50 ml glucose meals, neuronal responsiveness in the secondary taste cortex (caudolateral orbitofrontal cortex) decreased to the taste of the glucose solution but not to the taste of blackcurrant juice [Rolls et al., 1989]. Thus, not only can neurons of the secondary taste cortex distinguish glucose solutions from blackcurrant juice, but they can also associate the metabolic consequence of ingesting a glucose solution (satiety) with its specific taste. This responsiveness of orbitofrontal cortex neurons provides a mechanism for the anticipation of the nutritive value of a food substance before it is ingested.
Modern neuroimaging: fMRI
In recent decades more powerful tools have become available for the study of brain activity in living organisms. One of them is Functional Magnetic Resonance Imaging (fMRI). Scanning the brain with fMRI identifies changes in blood flow/oxygenation in various areas of the brain and such changes are used as proxy for neural activity. This non-invasive method is increasingly used to study neural control of appetite in humans. The anatomical resolution of images is steadily improving.
fMRI is a powerful tool capable of making fine discriminations between brain activities. For example, it was used in a study of neuro-economics in order to understand how the brain computes different value-related signals [Hare et al., 2008]. In order to make sound economic decisions, the brain has to perform multiple computations. “Goal values” measure the predicted reward associated with the outcome generated by each of several actions under consideration. “Decision values” measure the net value of taking the different actions (the benefits minus the costs). “Prediction errors” measure deviations from previous reward expectations. Goal values and decision values are used to guide decisions to those actions with the largest net benefit. Prediction errors are used to assess the value of states of the world and are critical for learning how to make optimal decisions in the future.
fMRI was used in a laboratory decision-making paradigm allowing researchers to dissociate the neural substrates related to the goal value, decision value, and prediction error computations [Hare et al., 2008]. Participants made decisions on 50 different sweet and salty junk foods presented in colour pictures, in terms of whether or not they were willing to purchase the food at a certain cost. Before entering the MRI scanner, the amount participants were willing to pay for each food was established, as an index of the reward value of each food. Once inside the MRI scanner, participants completed 300 trials of a forced-choice task using the same 50 food items with the price and the gain/loss randomly manipulated. The subjects responded with a button press to indicate whether or not they would pay the indicated price for the food item shown. Goal value (intrinsic value of the food), decision value (based on random prices), and prediction error could thus be largely dissociated, although not completely decoupled.
31