Much work on the neural substrates of habituation has been conducted on a group of marine invertebrates called Aplysia, the sea slugs (or sea hares), such as the species Aplysia californica shown in Figure 3.9. Like many marine animals, Aplysia breathes through gills, which extend upward from between the two wings of the mantle, the animal’s outer covering. A structure called the siphon works like a tube to blow aerated water over the gills to assist respiration. The gills are delicate and easily damaged. When danger threatens, the sea hare tends to retract them under the safety of the mantle. This is called a gill-withdrawal reflex.
One advantage of studying learning in Aplysia is that they have a relatively simple nervous system—only about 20,000 neurons, compared with the tens of billions in a cat or human. Plus, some of the neurons are very big. A few are
Figure 3.9 Aplysia californica, the sea hare This marine invertebrate, a shell-less mollusk, has a relatively simple nervous system, useful for study-ing the neural bases of learnstudy-ing. If the siphon is touched lightly, both siphon and gill are protectively withdrawn (the gill-withdrawal reflex). With repeated light touches, the gill-withdrawal reflex habituates. In this photo, the gill is underneath the mantle.
(Adapted from Squire and Kandel, 2000.)
Head Mantle
Siphon
Tail
David Wrobel/Visuals Unlimited
large enough to be seen with the naked eye. Best of all, the pattern of neurons in Aplysia seems to be “hardwired,” meaning that researchers can identify a particular neuron in one sea hare (say, motor neuron L7G) and find the same neuron in the same place in another member of the species. This type of nervous system makes things much easier for a neuroscientist trying to understand how the brain encodes new memories.
Neuroscientists have documented each of the neurons involved in Aplysia’s gill-withdrawal reflex. The siphon contains 24 sensory neurons that are directly connected to 6 motor neurons that innervate the gill. Figure 3.10a shows a simplified scheme of this system of neurons, consisting of three sensory neu-rons S, T, and U and one motor neuron M. When the siphon is touched, sen-sory neuron S fires, releasing a neurotransmitter, glutamate, into the synapse (Figure 3.10b). Molecules of glutamate diffuse across the synapse to activate receptors in motor neuron M. If enough receptors are activated, neuron M fires, causing the muscles to retract the gill for a few seconds.
As simple as sea hares are, they are still capable of adapting their behavior in response to experience. Aplysia show habituation, sensitization, and several other forms of learning, just as rats and humans do. In Aplysia, however, scientists can actually watch the nervous system in action as these learning processes occur.
Habituation in Sea Hares
An initial, light touch on a sea hare’s siphon will activate the gill-withdrawal response, but if the touch is repeated, the gill-withdrawal reflex gradually becomes weaker, or habituates. The amount of habituation is proportional to the intensity of the stimulus and the repetition rate. If a sufficiently light touch is delivered every minute, the withdrawal response habituates after 10 or 12 touches and can last for 10 to 15 minutes (Pinsker, Kupfermann, Castellucci, & Kandel, 1970).
In sea hares, we can see exactly what is causing this habituation. Refer back to the schematic diagram in Figure 3.10a. Recall that touching the siphon
Siphon Sensory neuron S
Tail Sensory neuron T
(a)
Mantle Sensory neuron U
Motor neuron M Gill muscles
S
Glutamate
M Glutamate receptor (b)
S
M
Fewer molecules released, so less transmission
(c) Figure 3.10 Neural
circuits in Aplysia gill- withdrawal reflex (a) Sensory neurons S, T, and U respond to a touch on the siphon, tail, and upper mantle, respectively. These neurons converge on motor neu-rons such as M, which contract the gill muscles. (b) When sensory neuron S fires, it releases the neurotransmitter glutamate into the synapse between S and M.
The glutamate molecules (shown in yellow) may dock at receptors on neuron M. If enough receptors are activated, neuron M will fire, retracting the gill. (c) If neuron S is activated repeatedly, it gradually releases less glutamate each time, decreasing the response of M.
This synaptic depression underlies habituation of the gill-withdrawal response in sea hares.
B R A I N S U B S T R A T E S
|
99excites sensory neuron S, which releases the neurotransmitter glutamate, which in turn excites motor neuron M, which drives the withdrawal response (Figure 3.10b). With repeated stimulation, however, neuron S releases less glutamate (Figure 3.10c), decreasing the chance that neuron M will be excited enough to fire (Castellucci & Kandel, 1974). The reduction in glutamate release is evident even after a single touch and lasts for up to 10 minutes. This decrease in neurotransmitter release is associated with a decrease in the number of glutamate-containing vesicles positioned at release sites. Thus, in Aplysia, habituation can be explained as a form of synaptic depression, a reduction in synaptic transmission. This is exactly the sort of weakening of connections proposed by the dual process theory of habituation (see Figure 3.3b).
An important feature of habituation in sea hares is that it is homosynaptic, meaning that it involves only those synapses that were activated during the habit-uating event: changes in neuron S will not affect other sensory neurons, such as T or U in Figure 3.10a. In other words, a light touch to the tail or upper mantle still elicits the defensive gill withdrawal, even though a touch to the siphon is ignored. Even the responsiveness of the motor neuron M is not changed; in this case, habituation in the short term affects only how much neurotransmitter neuron S releases.
Long-term habituation in sea hares can often last much longer than 10 minutes, especially when exposures are spaced over several days (Cohen, Kaplan, Kandel,
& Hawkins, 1997). How are the animals storing information about past exposures for such a long time? When a sea hare is repeatedly exposed to the same stimulus over several days, the actual number of connections between the affected sensory neurons and motor neurons decreases. Specifically, the number of presynaptic ter-minals in the sensory neurons of animals that have been repeatedly exposed to the same stimulus is reduced. Synaptic transmission in Aplysia can thus be depressed not only by decreases in neurotransmitter release but also by the elimination of synapses. This suggests that repeated experiences can lead not only to the weaken-ing of connections, as suggested by the dual process theory of habituation, but also to their elimination.
Do the mechanisms of habituation in sea hares tell us anything about habitu-ation in larger-brained animals? It is currently impossible to trace the entire neuronal circuit of habituation through the billions of neurons in a mammalian brain. However, neuroscientists have good reason to believe that the mechanisms of habituation documented in Aplysia occur in other species too. In fact, repeated stimulation of sensory neurons in other species, including crayfish and cats, also causes a reduction in neurotransmitter release. This suggests that at least some of the biological mechanisms of habituation are constant across species.
Sensitization in Sea Hares
What about sensitization, which, in contrast to habituation, causes increased responding to stimuli? Aplysia also provide a way to study the neural processes involved in this kind of learning. Suppose, instead of a light touch to the siphon, the researcher applies a more unpleasant stimulus: a mild electric shock to the tail that causes a large, sustained gill-withdrawal response. The aversive tail shock sensitizes subsequent responding, so that a weak touch to the siphon now produces a strengthened gill withdrawal.
To understand how this occurs, let’s take the simplified circuit diagram from Figure 3.10a and add one more level of neural detail, as shown in Figure 3.11a.
The tail shock activates sensory neuron T, which activates motor neuron M, causing the gill-withdrawal response. But because of the arousing nature of shocks, neuron T also activates modulatory interneurons, such as IN. An
interneuron, as its name suggests, is a neuron that neither directly receives senso-ry inputs nor produces motor outputs but instead carries a message between two other neurons. A modulatory interneuron is an interneuron that alters the strength of the message being transmitted. You’ll recall from Chapter 2 that neuromodula-tors are neurotransmitters that can affect activity in entire brain areas rather than just at a single synapse. In Aplysia, interneuron IN connects neuron T to both S and U, communicating with them by releasing a neuromodulator such as sero-tonin. Serotonin increases the number of glutamate vesicles available to release glutamate from neuron S each time it fires. In effect, the interneuron does not tell S whether to fire; instead, it tells S, “When you do fire, fire strongly.”
Suppose the sea hare now experiences a mild touch to the siphon. Before the tail shock, this mild touch would have caused neuron S to release neurotrans-mitter from a small number of vesicles, leading to a weak gill withdrawal (as in Figure 3.10b). But now, this mild touch causes neuron S to release neurotrans-mitter from a larger number of vesicles (Figure 3.11b), so that M is more likely to fire, leading to a stronger gill-withdrawal response (Brunelli, Castellucci, &
Kandel, 1976; Castellucci & Kandel, 1976). In effect, the prior tail shock at T has put the slug on alert, making it sensitive to a subsequent light touch.
The key to sensitization is that it is heterosynaptic, meaning that it involves changes across several synapses, including synapses that were not activated by the sensitizing event. Because of this feature, a tail shock increases responses to any future stimulus. For example, if the experimenter touched the upper mantle (activating sensory neuron U) instead of the siphon (neuron S), the same over-reaction would occur.
In effect, the tail shock has increased the sea hare’s level of arousal, making it more likely to respond to any stimulus that follows. As with habituation, the degree of sensitization depends on the strength of the initial stimulus: a single mild shock can produce sensitization that lasts for minutes; four or five shocks together can produce sensitization that lasts two or more days (Marcus et al.,
(a) IN
Siphon Sensory neuron S
Tail Sensory neuron T
Mantle Sensory neuron U
Sensory neuron M Gill muscles
S
More vesicles, so more transmission
M (b) Figure 3.11 Sensitization in Aplysia
(a) A tail shock activates sensory neuron T, which activates motor neuron M, causing the motor response. T also activates an interneuron IN, which delivers a neuromodulator (such as serotonin) to the axons of neurons S and U. (b) Subsequent activation of neuron S will cause a larger release of neurotransmitter (glutamate), leading to greater activation of neuron M than was previously evoked by S.
B R A I N S U B S T R A T E S
|
1011988; Squire & Kandel, 2000). Note that in the sea hare, sensitization does not engage a separate state system, as suggested by the dual process theory of habituation. Instead, interneurons that are part of the circuit that generates the reflexive response increase synaptic transmission. Nevertheless, two indepen-dent processes are associated with habituation and sensitization effects, and so the dual process theory of habituation does a good job of accounting for these effects in sea hares. Dual process theory was originally developed to explain habituation in mammals, so the fact that it can also capture many of the aspects of habituation and sensitization in a marine invertebrate suggests that many different species learn about repeated events in similar ways.
Test Your Knowledge
Synaptic Mechanisms of Learning in Aplysia
Habituation and sensitization have different effects on synaptic transmission in sea hares. For the following situations observed in sea hares, see if you can deduce whether habituation or sensitization has occurred and what is happening in the stimulus-response pathway. (Answers appear in the Answers section in the back of the book.)
1. Observations reveal that only one synapse, where once there were two, is now connecting a neuron that responds to stimulation of the tail to a neuron that contributes to the gill-withdrawal response.
2. Measurements of glutamate released around a motor neuron show that levels of glutamate are increasing over time.
3. Anatomical analyses of neural circuits reveal a larger number of synapses associ-ated with the gill-withdrawal reflex circuit than are normally seen.
4. Recordings of motor neuron activity indicate that the neurons are firing less.