Memory in Fishes – Fact or Fiction, by Thomas Schubert

Ask just about anyone and they will tell you that goldfish can’t remember much more than about ten seconds. However, this response raises several questions. How do fish remember where the best cover is? How do fish remember what prey source is the most energetically efficient? Why is it I can only catch little fish (the big ones can’t possibly have seen my lure before)? The latter being the most frustrating. It is obvious after some consideration that this old story can’t possibly be true. While some argue fish are incapable of long term memory, fish do in fact demonstrate the ability to remember, which can be understood through scientific observation and experimentation.

Most heavily pressured populations of fish become somewhat harder to catch for recreational anglers. This is likely due a recognition of common lure patterns, or perhaps a correlation made between a person on the bank and a higher chance of harm. Regardless, both scenarios demonstrate some form of long term memory in the fish. One study tested this theory in brown trout (Young et al. 2003). In this study researchers fished two rivers in New Zealand, the Ugly River and the Owen River. The Ugly River is very remote and receives little angling pressure, while the Owen River is heavily fished. After fishing both rivers on four separate occasions, up to 47% of the trout population of the Ugly River were caught, yet only 11 and 23% percent of the population of the Owen River were caught. Additionally, on the Ugly River, it was found that trout that had been previously caught were rarely seen after the occurrence. This did not occur on the Owen River. This behavioral difference between the two rivers may suggest and impact of fear, as well as learning. On the Ugly River, where fish had not been exposed to anglers, fear after being caught likely caused the fish to seek heavy cover, and not be seen again. Whereas on the Owen River, where fish have experienced a great deal of angling pressure, the Brown Trout acted in a different manner. Rather than exhibiting a fear response, fish that had been previously caught just did not bite again. This lower susceptibility to common lures exhibits a learned behavior in the Brown Trout of heavily pressured populations.

Brown Trout being released to a New Zealand River.  Source 

Another profound study covering memory in fish was performed on Goldfish (Davis et al. 1966). In this study Goldfish were placed in an electric “shuttle box”, seen in Figure One below, and exposed to electric shock avoidance training 20 times during a 40-minute span. Fish exhibited an increase in avoidance behavior after receiving the shocks. Furthermore, some fish were shocked again three days later. These fish exhibited even stronger avoidance behaviors. Interestingly, another group of fish were tested in the same manner, but were left in the shuttle box for an hour after training (instead of being brought back to their home tank). In this group it was found that the memory at three days was somewhat poor, and not much better than an untrained fish. This suggests that high stimulation and stress impedes the ability of fish to create long term memories. This implies that fish in their natural environment, often under high stress, would lack in memory formation. In my understanding, this is somewhat analogous to cramming for a test, while it may work temporarily, the stress and lack of repetition leads to poor long term memory.

Figure 1 Illustration of the "shuttle box" used to perform shock avoidance training, this is an overhead view (Travino, 1969)

As compelling as this study already is, the researchers went on to further solidify their point. They did so by injecting the fish with puromycin, an antibiotic that inhibits protein synthesis. Fish were injected immediately following training, and it was found that three days later there was no recollection of the shock training (the fish tested the same as first time avoidance training). The researchers then tested fish injected immediately after training at shorter time spans, typically one and two days. It was found that in these fish there remained a somewhat strong memory of the shock training, suggesting a form of short term memory. However, these same fish exhibited little to no memory at 72 hours. This study suggests that long term memory is the result of a structural change to the brain that occurs sometime after training, whereas short term memory is not. Other studies have corroborated this finding, and go on to find that consolidation of memories takes place somewhere between 90 and 100 minutes after training (Travino, 1969). This experiment also suggests that short term memory in fish persists for approximately 48 hours, and after this period any learned behavior is a result of long term potentiation. While this study does not delve into memory formed much past three days, other studies have, and suggest that goldfish form memories that can persist for more than six months (Handler, 2015). This process of memory formation is similar in most every vertebrate, including humans, suggesting that some of our basic abilities to learn may be somewhat ancestral.

It is apparent through scientific observation and experimentation that fish demonstrate the ability to remember, and in turn, learn. However, the extent to which they can perform this ability in the wild is unclear. Rather, the answer probably lies somewhere in between. That is, the ability to learn is likely dependent upon environmental factors and stressors, the particular species, as well as the individual within that species. Further experimentation in the wild would be necessary to determine the extent to which fish express the ability to remember in their native environment. It is apparent though that the ability does in fact exist. Or perhaps it is only me that they remember.

References

Davis, R. E., and B. W. Agranoff. 1966. Stage of Memory Formation in Goldfish: Evidence for an Environmental Trigger. Proc Natl Acad Sci U S A.  55(3): 555–559.

Handler, K. (n.d.). Goldfish as a Model for Understanding Learning and Memory: More Complex Than You Think. American Museum of Natural History.

Travino, J. E. 1969. The Effect of Puromycin and Electroconvulsive Shock On Retention of Shock Avoidance in the Goldfish. thesis.

Young, R. G., and J. W. Hayes. 2003. Angling Pressure and Trout Catchability: Behavioral Observations of Brown Trout in Two New Zealand Backcountry Rivers. North American Journal of Fisheries Management 24(4).