Everyone Can Swipe Right — Exploring the use of Touch Screen Technology to Test Specific Cognitive Functions in Mice.
Touch screen devices threaten to be the preferred cross-species learning tool of the future. It’s kind of awesome.
*** Author’s note — Upon re-reading this post, I feel like I get way too ‘Sciency in some places. If this puts you off, please drop me a line. Regardless, I plan to come back to it shortly to make it a little bit more accessible.
The complexity of the mammalian brain is in part exhibited in the collaborative and independent functioning of its various core components (see Figure 1). So compartmentalized is the brain, that a mammal can sustain severe head trauma in one area, and find that not all brain functions are affected. The Hippocampus (HPC) and Dorsal Striatum (DS) have been heavily studied because they play complimentary roles in a variety of brain functions (memory, learning, etc), but are functionally dissociated. New research suggests that impaired functioning in the hippocampus does not adversely affect the acquisition of new skills learned through touch screen devices; this is presented in contrast to a remarkable decrease in one’s ability to learn through the exploration of a physical environment (Delotterie et. al, 2015).
Figure 1. Basic brain components, including the Hippocampus, Striatum, Ventral Tegmental Area (VTA), and others. ** I forget where I got this image; if it’s yours, let me know and I’ll site you properly.
Delotterie et. al, (2015) present evidence that complex tasks can be taught to subjects after severe HPC damage using touch screen devices, where the same task becomes “unteachable” through less technologically advanced methods. Furthermore, both the HPC and the Striatum are involved in multiple types of decision-making, with dorsolateral striatum involved in stimulus-response strategies and ventral and dorsomedial striatum involved in goal-directed strategies (Johnson et. al., 2007).
I think the implications here are boundless.
Delotterie et. al (2015) examined the independence of the HPC and DS by applying lesions to those brain areas in mice before and after new skills had been acquired. They found that lesions applied to the DS before skill acquisition (pre-training) caused significant impairment regardless of the learning and testing mechanism. The impact of HPC lesions was much more nuanced. Pre-training lesions resulted in impaired practical learning, but allowed for the acquisition of new skills through a touch screen device. Post-training HPC lesions universally impacted the recall of previously learned tasks (Delotterie et. al, 2015).
Dellotterie et. al (2015) conceived of two experiments measuring task assimilation and recall (experiments A and B). Experiment A began with either real or fake excitotoxic (destruction through hyperstiumlation) lesions to either the HPC or DS in all participants (in this case, 54 four-month old mice). Every mouse received an injection; experimental group needles containing a chemical solution that would cause acute neuronal destruction in targeted areas (the lesions), whereas control groups were injected with an inert agent. Four weeks after surgery, learning tasks began. The task set included a combination of practical and theoretical learning paradigms. A Paired Associates Learning (dPAL) task (like flipping over two cards from a set, and putting matching pairs together), and a Visuo-Motor Conditional Learning (VMCL) task (observing a visual representation of another mammal performing a behaviour) were both delivered via touch screen device.
Mice also performed continuous alternation tasks that observed their performance as they explored a simple maze. Experiment B saw the intake of a new batch of mice who were trained in the dPAL task and subsequently lesioned in the HPC and DS (or not, in the control group). Once allowed to recover, the mice in Experiment B were re-tested in the T Maze to gauge the impact of the lesions on their memory of the maze. All tasks were scheduled to be taught over the course of eight five-block training sessions (Delotterie et. al, 2015).
Experiment A featured a gradual performance improvement in most participant groups (mice were divided into one group for each lesion type, and each lesion type had an associated control group). Performance of animals featuring HPC lesions (real or fake) did not differ from each other. Conversely, in mice subjected to DS lesions, there was almost no improvement (RM 2-way ANOVA: F(21,266) = 4.39; p < 0.0001). At the beginning of the fourth 5-session block, performance of mice with DS lesions was significantly below that of the three other groups (Tukey’s post hoc tests: all q > 4; all p < 0.05), leading researchers to conclude that this group would not be successful in learning the new task in the allotted time (see figure 2 for details). A similar pattern was observed in the performance during the VMCL task (Delotterie et. al, 2015).
Figure 2. New task performance improvement observed over the course of prescribed training sessions (Delotterie et. al, 2015).
Performance was approximately equally impaired in the T Maze across both lesion-types. Where HPC and DS sham controls performed similarly (Tukey’s post hoc test: q(38) = 0.50; p = 0.985), a significant difference was found between HPC lesioned and HPC sham groups on the one hand (Tukey’s post hoc test: q(38) = 6.82; p = 0.0001), and between DS lesioned and DS sham groups on the other hand (Tukey’s post hoc test: q(38) = 4.33; p = 0.020). Experiment B featured no major anomalies, in that all groups absorbed new tasks equally well, and experimental groups were equally impaired by lesions (Delotterie et. al, 2015).
It is currently unclear how vividly the Delotterie et. al (2015) results would translate into specific Decision-Making tasks. The HPC and Striatum are both intimately associated with memory retention, storage, and reactivation; where the HPC is linked to the provision of context through short term memory cues (Johnson et. al., 2007), the Striatum is more involved with reinforcement and future planning through the release/retention of dopamine (Wickens et. al, 2007).
Noting that neuronal firing rates are correlated with integration of sensory evidence used to inform decision-making, (Jahans-Price, et. al., 2014), the interruptions in the learning process associated with striatal damage would likely yield similar trends in decision tasks; it is unclear as to how a damaged HPC would respond. Regardless, with this information in hand, one might genuinely muse that that hospitals will soon feature an iPad (or android equivalent) at every bedside, if not exclusively in areas reserved for patients suffering from neurological damage, to facilitate rehabilitation.
References
Delotterie, D. F., Mathis, C., Cassel, J. C., Rosenbrock, H., Dorner-Ciossek, C., & Marti, A. (2015). Touchscreen tasks in mice to demonstrate differences between hippocampal and striatal functions. Neurobiology of Learning and Memory, 120, 16–27.
Jahans-Price, T., Gorochowski, T. E., Wilson, M. A., Jones, M. W., & Bogacz, R. (2014). Computational Modeling and Analysis of Hippocampal-Prefrontal Information Coding During a Spatial Decision-Making Task. Frontiers in Behavioural Neuroscience, 8.
Johnson, A., van der Meer, M. A., & Redish, A. D. (2007). Integrating Hippocampus and Striatum in Decision-Making. Current Opinion in Neurobiology, 17(6), 692–697.
Wickens, J. R., Budd, C. S., Hyland, B. I., & Arbuthnott, G. W. (2007). Striatal Contributions to Reward and Decision Making. Annals of the New York Academy of Sciences, 1104(1), 192–212.