Learning; it is considered the foundation on which organisms can change their behaviour beyond basic instinct. The benefits of being able to change one’s behaviour in response to something are far reaching and can improve fitness and biological success. This paper tested the response of the Polychaeta genus Seripulidae to repeated stimulus, designed to mimic ecological scenarios like predation. The results showed that there was a correlation between the times take to protrude from their tube between repeated stimuli. This indicated that there was a significant change due to the stimulus, which is believed to be indicative of learning. 

Across the animal kingdom, instinct serves individuals well; guiding them towards survival through neural pathways developed over millions of years of biological development. Instinct however is only a truly basic survival tool. The ability to learn and respond differently to situations and stimulus gives many animals an advantage over those purely reliant on instinct. Learning is widely accepted as the alteration of synapse potency within pathways that mediate behaviour (McClellend & Rummelhart,1986), a concept that is founded on Ramon Y Cajal’s study from 1894. Cajal’s paper paved the way for twentieth century technology to reveal more about synapse plasticity in learning. For much of the twentieth century however literature regarding plasticity focused on localised synapses (Wilson &McNaughton, 1993). Competing theories now suggest that particular types of learning may be induced through the modulation of higher nervous centres (Leaton & Supple, 1986).  Kranse& Teshiba (1995) have even presented evidence that habituation of invertebrates does occur solely due to extrinsic modulation.  This is an interesting observation as it suggests more complex neural function in invertebrates may be common. The study of this complexity may inform researchers about the ecological benefits of learning in simple annelids as well as provide data to help understand specific species phylogeny through more detailed neural study (Brinkmann &Wanninger, 2008).

The subject invertebrates of this study are comprised of various species of the Seripulidae genus. Being simple tubeworms they display common instinctive reflexes, such as an escape reflex when stimulated by predators. This reflex is thought to be a result of synapse reaction along a large longitudinal nerve cord. Observations have led me tohypothesise that these calcareous tubeworms increase time between protrusions of their tentacle crown, with exposure to repeated stimulus. It is my theory that repeated stimulus is will change the time between protrusions of the Seripulidae and thus will demonstrate an ability to learn. 

Varying worms of the Seripulidae genus were chosen from growth plates taken from Manly harbour in Moreton Bay. Each worm used was housed on its own individual growth plate (approximately 10cm x10cm). The Seripulidae were stimulated three times to test for an increase in time between retractions until when it next fully protrudes its tentacle crown(protrusion period). The tests were conducted immediately, one after the other. Each stimulus was only conducted once the fan is fully extroverted. Stimulus was provide by lightly stroking the crown with metal tweezers. Each plate was submerged in small, white plastic containers while the trials were conducted. Trials were conducted under ambient lab light and care was taken not to cast shadows across the plates while experiment was underway, so as to not compromise the repeated reactions. 

A correlation between the repeated stimulus and time taken to protrude from tubes was determined using multiple regression.Treating the third protrusion period (s3) as the explanatory variable, the first two protrusion periods (s2 and s1) were tested for correlation. It was clear however that results were giving irregular residuals; therefore a log transformation of the data set was used to establish correlation.  The Q-Q plot (Graph 1) and residual plot(Graph 2) both suggest the data in normally distributed, save a couple of outliers. A summary of coefficients of the log transformation then indicates a significant level of correlation between the second stimulus and the third (Figure1; p=0.04). On top of this, the adjusted R² states that the log model can explain approximately 62% of the variation. 

Table 1

Summary of the Log Coefficients

Coefficients:

                                 Estimate       Std. Error      t value       Pr(>|t|) 

(Intercept)             -1.1207        1.6160           -0.693         0.4925 

log(s2)                     1.1170         0.5289           2.112         0.0417*

log(s1)                     0.8333         0.4634            1.798        0.0806 .

log(s2):log(s1)      -0.1437         0.1298           -1.107         0.2757 

Residual standarderror: 0.6088 on 36 degrees of freedom

MultipleR-squared:  0.6527, AdjustedR-squared:  0.6238

F-statistic: 22.55 on3 and 36 DF,  p-value: 2.162e-08

Figure 1 shows a significant correlation between the time recorded for the second and third protrusion period (t-value=2.1,p=0.04). No other significant results were recorded. The adjusted R² value shows the model explains approximately 62% of the variation. 

The results indicate a correlation between the time taken by the Seripulidae worms to fully protrude after the second and third stimulus, with a p-value of 0.04 (Table 1). This result is validated in the data by a consistent Q-Q plot (Figure 1) and a symmetrical residual plot (Figure 2). Though this result does not provide directionality, it gives an indication that the second stimulus affects the last protrusion period. I believe this suggests that the worms are altering their reaction significantly, or learning from one stimulus to the next. The ability to learn or even habituate to an environment has been seen in annelids before. Clark (1960), notes in his study that individual Nereis pelagica, a polychaete of the Nereididae family, can react differently to stimuli but most do alter their reaction to repeated stimulus. His work showed however that such tube annelids could exhibit habituation by ignoring stimulus completely: a phenomenon known as stimulus filtering (Simmons & Young, 1999). This result is most likely the effect of neurons not firing to the ventral nervous cord which runs laterally along the body and is the source of the withdrawal mechanism in tube worms (Ruppert, Fox& Barnes, 2004). 

Though the Seripulidae worms did not show signs of stimulus filtering, as a demonstration of learning, they did show that there are changes in neural activity, due to the correlation between the second and third stimulus protrusion periods (Table1). The fact that there was no sign of stimulus filtering would suggest that synaptic pathways to the giant longitudinal axon of the worms were not altered. However the significant change in time between the second and third protrusion period suggests an alteration of synapse potency in the brain, or similar higher neural system (Kranse & Teshiba, 1994). This is because activity beyond simple instinct is generally associated with higher nervous function. Though it is also been mentioned in several pieces of literature that invertebrate neuroplasticity may arise through a cascade of molecular events in multiple systems (Shaw, Lanius & Doel, 1994), rather than just through modulation of the brain. This idea is given credence by the ladder like construct of annelids, in which each segment ganglia are consider smaller brains (Ruppert, Fox & Barnes, 2004).  Regardless of the pathway however I think it may not be to bold to state that simple tubeworms, such as the Seripulidae, possess the ability to react to stimulus, or learn, in ways outside of simple instinctive reflexes. Furthermore, the abundance of literature on polychaete learning and habituation helps to cement my confidence in this conclusion.

Keeping in mind the limited trials of the experiment, it is assumed that permanent habitual behaviour of the worms would not occur. This would suggest that non-synaptic plasticity could explain the difference in stimulus protrusion periods, as a change in the chemistry of synapses is not necessarily permanent. (Kemenes et al., 2006).  Yet, Kemenes (2006) notes that modulatory neurons could non-synaptically influence long-term memory. Of course, the subject of this experiment was Lymenaea Stagnalis, a gastropod, and so the memory capacity may differ with the Seripulidae worms. In saying this though, Kemenes’ work could be a good model for future study on learning in annelids, particularly neural pathways.

Considering this evidence for learning in Seripulidae worms, it is possible to recognise several ecological benefits for such animals. With regard to stimulus, defence mechanisms are one of the most obvious of issues that could be improved through learning and, if sustained for long enough, habituation. For example, the tubeworms increase in time between protrusions could act to increase their fitness by increasing time behind the operculum (Ruppert, Fox & Barnes, 2004), a protective tubeplug. This would limit their exposure to large natural predators. Adapting such defence mechanisms may not only protect Seripulidae from large macro-predators but also small microorganisms, a particularly concerning issue in farmed invertebrate communities (Roch, 1999). Naturally, responses to microorganisms would occur more within the chemistry of the animal in question. However the same benefits of neural plasticity can apply to both physical and chemical defence.

On top of predation, exposure to environmental factors may also reveal forces that dictate learning within invertebrates such as the Seripulidae. Research from Hwang, Costello and Strickler, (1993) on the copepod Centropages hamatus, indicated changes in foraging patterns between turbulent and non-turbulent water conditions. The study created opportunity for habituation. This is something that, if conducted within my own experiment,would have helped determine not only an ability to learn, but also an ability to incorporate learning permanently into the life of individual Seripulidae worms. Such an extension of my experiment may also reveal evidence of physical changes in synapse pathways as required by synaptic plasticity (Kemenes et. al, 2006). This would be in contrast to how learning is achieved on a more temporary basis.

On top of the ecological comprehension and the sheer curiosity of understanding learning, benefits may also be discovered in phylogeny. The more complex issue of neurogenesis in larval polychaetes was examined by Brinkmann and Wanniger (2008), to help in creating a better understanding of metazoan evolution. Body segmentation in particular is one phylogenetic aspect that needs attention.Though such work is far reached from my own experiment, I believe that moving towards more complex analysis of neural systems in polychaetes such as Seripulidae is invaluable. What’s more, even using behavioural response, such as those demonstrated in my experiment, have been used to help determine phylogeny before (Walter & Proctor, 1998), though not necessarily on annelids.

Overall, this experiment has warranted support of the proposed hypothesis. Despite its rudimentary nature and time constraints, I believe the significant correlation between the second and third protrusion periods demonstrates a learning ability. As the experiment was conducted on such a short time scale, it is assumed that habitation did not occur and that non-synaptic pathways were most likely how they responded. Though, the specific areas in which chemical or physical changes could occur is still up for debate in the literature. Plus, it is made more complex by the nerve structure of polychaetes like Seripulidae.  Regardless, the ability to learn in response to stimulus could prove useful in many ecological scenarios and, upon closer study of the nervous system, could provide useful information for helping determine phylogeny. The most natural pathway for further study in this area would be test for habituation in response to stimulus. There are many concepts within this paper though which merit investigation beyond my own undergraduate comprehension. 

Alcock, J. (2009). Reaction to stimulus and fitness - Animal Behavior (Ninth ed., Vol. 1). Sunderland, MA: Sinauer Associates, Inc.

Brinkmann, N. & Wanniger, A. (2008). Larval Neurogenesis in Sabelleria alveolata reveals plasticity in polychaete neural patterning. Evolution & Development. 10(5): 606-618.

Hawkins, R.  D.,Kandel, E. R. & Siegelbaum, S. A. (1993). Learning to modulate transmitter release: Themes and variations in synaptic plasticity. Annual review of Neuroscience,16, 625-665.

Hwang, J., Costello, J. & Strickler, J. (1993). Copepod grazing in turbulent flow: elevated foraging behavior and habituation of escape responses. JPR. 16(5):421-431. doi: 10.1093/plankt/16.5.421

Kemenes, I., Straub, V., Nikitin, E., Staras, K., O'Shea, M., Kemenes, G. & Benjamin, P. (2006). "Role of Delayed Nonsynaptic Neuronal Plasticity in Long-Term Associative Memory". Current Biology 16(13): 1269–1279. doi:10.1016/j.cub.2006.05.049.PMID 16824916.

Kranse, F. & Teshiba, T. (1995). Habituation of an invertebrate escape reflex due to modulation by higher centers rather than local events. PNAS. 92(8):3362-3366. doi: 10.1073/pnas.92.8.336 

Leaton, R. N. & Supple, W.  F., Jr. (1986). Cerebellar vermis: essential for long-term habituation of the acoustic startle response.  Science. 232, 513-515.

Roch, P. (1999). Defence mechanisms and disease prevention in farmed marine invertebrates. Elsevier.172(1-2): 125-145.

Rumelhart, D. E., & McClelland(1986). On learning the past tenses of English verbs. In McClelland, J. L.,Rumelhart, D. E., and the PDP research group (Eds.) Parallel distributedprocessing: Explorations in the microstructure of cognition. Volume II.Cambridge, MA: MIT Press. Chapter 18, pp. 216-271.  

Ruppert, E., Fox, R. & Barnes, R. (2004). Invertebrate Zoology: A functional evolutionary approach. 7^th Edition. Brooks/Cole. Belmont.

Shaw, C., Lanius, R. & Doel, K. (1994). The origin of synaptic neuroplasticity: crucial molecules or a dynamical cascade? Elsevier. 19(3): 241-263.

Simmons, P. & Young, D. (1999). Stimulus filtering: vision and motion detection. In: Nerve Cells and Animal Behaviour.pp. 99-128. [Online]. 2^nd ed. Cambridge University Press.

Walter, D. & Proctor. H. (1998).Feeding behavior and phylogeny: observations on early derivative Acari. Springer. 22(1): 39-50

Wilson, M. A. & McNaughton, B. L. (1993). Dynamics of the hippocampal ensemble code for space. Science. 261, 1055-1058.