PSYC202 Foundations of Cognitive Neuroscience – Week 4: Motivation, Reward, and
decision-making
Like all other animals, there are certain behaviours that we need to engage in to
survive. Specifically, we must seek and interact productively with aspects of our environment
that are necessary for our survival, food and water being the most obvious examples, and we
must successfully avoid harmful situations. Taking an evolutionary perspective (LeDoux,
2012; Rolls, 2014) we can see that reflexive behaviours can go some limited way towards
achieving these necessary interactions with the environment. A frog with a reflex to quickly
aim a long tongue when a moving, fly-shaped object is in the vicinity will achieve some
nourishment. Reflex systems, by their nature, are rigid entities however, and allow very little
flexible adaptation to changing conditions; if the supply of local flies becomes scarce in the
frog’s environment due to weather conditions for example, the tongue reflex in and of itself
will not aid survival. More flexible systems evolved whereby stimuli in the environment that
deliver primary rewards, link to brain systems that induce learning to occur, so that
previously relatively reward-neutral stimuli can come to be sought, or avoided, because they
deliver reward or punishment respectively. In the words of Berridge and Kringelbach (2015,
p.646), “…pleasure can be thought of as evolution’s boldest trick…”. The term primary
reward refers to stimuli that we find innately rewarding, whereas secondarily reinforcing
stimuli are stimuli that were initially neutral but became sought-after following learning, if
they were associated with primary reward. Thus, we can think of brain systems that allow
primary rewards to serve a teaching function, allowing us to orient to aspects of the
environment that will lead to rewarding outcomes. In this lecture we will focus on reward
systems, as opposed to punishing environmental stimuli, the questions being, how reward is
signalled in the brain, how this reward can cause learning to occur (usually termed
reinforcement learning), and how reward “value” influences decision-making in action
selection.
Primary rewards and “value”
Which aspects of the environment serve as primary rewards (sometimes also called
primary reinforcers)? This is a surprisingly hard question to answer fully, although some
stimuli incur more general agreement than others. Rolls (2014, p.20) suggests a
categorisation of stimuli in terms of rewarding tastes (sweet, salty and umami, the taste of
protein), odours, somatosensory (e.g. touch, grooming, optimal temperature-seeking, visual)
and auditory (soothing vocalisations, music). He also lists reinforcers linked to reproduction,
and then a set of reinforcers that seem to represent complex psychological systems
themselves, such as intellectual problem solving, desire for novelty, exercise etc. Perhaps the
truth about some complex reinforcers is that there is a large degree of individual variation in
people’s spontaneous preferences (e.g. reinforcement value of music, or being in nature as
opposed to the built environment), but this does not necessarily detract from the argument
that some of these preferences may be spontaneous or unlearned.
How is reward represented in the brain? Are there a relatively small set of brain areas,
or types of neurochemicals that are active following rewarding stimuli? Are these systems
common to different types of rewarding stimuli, such as taste, sound or touch, or are there
diverse systems for differing classes of rewarding stimuli? The early answer to this question
was that there were relatively few reward centres (or “pleasure centres”), and that these
centres were rich in neurons that released the neurotransmitter dopamine. The first studies
were conducted by Olds and Milner (1954; Olds, 1956) on rats, using the then newly
invented technique of electrode stimulation. A thin wire with two tips is inserted into a
1
decision-making
Like all other animals, there are certain behaviours that we need to engage in to
survive. Specifically, we must seek and interact productively with aspects of our environment
that are necessary for our survival, food and water being the most obvious examples, and we
must successfully avoid harmful situations. Taking an evolutionary perspective (LeDoux,
2012; Rolls, 2014) we can see that reflexive behaviours can go some limited way towards
achieving these necessary interactions with the environment. A frog with a reflex to quickly
aim a long tongue when a moving, fly-shaped object is in the vicinity will achieve some
nourishment. Reflex systems, by their nature, are rigid entities however, and allow very little
flexible adaptation to changing conditions; if the supply of local flies becomes scarce in the
frog’s environment due to weather conditions for example, the tongue reflex in and of itself
will not aid survival. More flexible systems evolved whereby stimuli in the environment that
deliver primary rewards, link to brain systems that induce learning to occur, so that
previously relatively reward-neutral stimuli can come to be sought, or avoided, because they
deliver reward or punishment respectively. In the words of Berridge and Kringelbach (2015,
p.646), “…pleasure can be thought of as evolution’s boldest trick…”. The term primary
reward refers to stimuli that we find innately rewarding, whereas secondarily reinforcing
stimuli are stimuli that were initially neutral but became sought-after following learning, if
they were associated with primary reward. Thus, we can think of brain systems that allow
primary rewards to serve a teaching function, allowing us to orient to aspects of the
environment that will lead to rewarding outcomes. In this lecture we will focus on reward
systems, as opposed to punishing environmental stimuli, the questions being, how reward is
signalled in the brain, how this reward can cause learning to occur (usually termed
reinforcement learning), and how reward “value” influences decision-making in action
selection.
Primary rewards and “value”
Which aspects of the environment serve as primary rewards (sometimes also called
primary reinforcers)? This is a surprisingly hard question to answer fully, although some
stimuli incur more general agreement than others. Rolls (2014, p.20) suggests a
categorisation of stimuli in terms of rewarding tastes (sweet, salty and umami, the taste of
protein), odours, somatosensory (e.g. touch, grooming, optimal temperature-seeking, visual)
and auditory (soothing vocalisations, music). He also lists reinforcers linked to reproduction,
and then a set of reinforcers that seem to represent complex psychological systems
themselves, such as intellectual problem solving, desire for novelty, exercise etc. Perhaps the
truth about some complex reinforcers is that there is a large degree of individual variation in
people’s spontaneous preferences (e.g. reinforcement value of music, or being in nature as
opposed to the built environment), but this does not necessarily detract from the argument
that some of these preferences may be spontaneous or unlearned.
How is reward represented in the brain? Are there a relatively small set of brain areas,
or types of neurochemicals that are active following rewarding stimuli? Are these systems
common to different types of rewarding stimuli, such as taste, sound or touch, or are there
diverse systems for differing classes of rewarding stimuli? The early answer to this question
was that there were relatively few reward centres (or “pleasure centres”), and that these
centres were rich in neurons that released the neurotransmitter dopamine. The first studies
were conducted by Olds and Milner (1954; Olds, 1956) on rats, using the then newly
invented technique of electrode stimulation. A thin wire with two tips is inserted into a
1
