Dissertation committee:
1. Μουτούσης Κωνσταντίνος, Καθηγητής, Τμήμα Ιστορίας και Φιλοσοφίας της Επιστήμης, ΕΚΠΑ.
2. Σαμαρτζή Σταυρούλα, Καθηγήτρια, Τμήμα Ψυχολογίας, Πάντειο Πανεπιστήμιο Κοινωνικών και Πολιτικών Επιστημών.
3. Βατάκη Αργυρώ, Αναπληρώτρια Καθηγήτρια, Τμήμα Ψυχολογίας, Πάντειο Πανεπιστήμιο Κοινωνικών και Πολιτικών Επιστημών.
4. Σκαλιώρα Ειρήνη, Καθηγήτρια, Τμήμα Ιστορίας και Φιλοσοφίας της Επιστήμης, ΕΚΠΑ.
5. Ζιώρη Ελένη, Επίκουρη Καθηγήτρια, Τμήμα Ψυχολογίας, Πανεπιστήμιο Ιωαννίνων.
6. Μέλλον Ρόμπερτ, Καθηγητής, Τμήμα Ψυχολογίας, Πάντειο Πανεπιστήμιο Κοινωνικών και Πολιτικών Επιστημών
7. Αβρααμίδης Μάριος, Καθηγητής, Τμήμα Ψυχολογίας, Πανεπιστήμιο Κύπρου
Summary:
Time perception is fundamental to our experience and central to virtually all of our everyday activities. Several studies have revealed an improvement in temporal discrimination when temporal perceptual learning was applied, implying that our temporal percept is plastic, depending on both experience and other cognitive/neuronal factors (Bueti et al., 2012). However, in most of the previous studies where a unisensory stimulation was applied, the learning effects could not be transferred to untrained intervals (e.g., Karmarkar & Buonomano, 2003) or modalities (e.g., Lapid et al., 2009), while the few studies that have attempted to approach learning through multisensory events, observed discrepancies in the transfer of learning from an asymmetric generalization (e.g., Alais and Cass 2010) up to none (e.g., Virsu et al., 2008).
Among the core methodological issues of the existing literature as described in Chapter 2, was the single temporal task applied both as the learning and as the evaluation task. Through this manipulation, the participants were repeatedly exposed to the same task for an exhaustive number of trials, suggesting that any induced learning effect most likely reflected a familiarization effect rather than a “pure” time processing improvement. Moreover, the non-engaging abstract stimulation (i.e., lacking any semantical content and a challenging purpose) might not give a clear image of the time perception changes or even induce a temporal learning of generalization effect to its full potential. In this thesis, therefore, the aim was to develop an alternative methodological tool to investigate, for the first time, how temporal perceptual learning through unisensory and multisensory events that carry a semantical meaning (i.e., Morse code language learning) could improve temporal perception skills.
The adoption of the Morse code as the main methodological approach was based on the plethora of benefits that it could offer, as described in full in Chapter 3. Its amodal and redundant character ensured an equal learning task in terms of information for each modality and its multisensory adaptation, allowing a direct comparison between the unisensory and multisensory conditions. Most importantly the nature of the Morse code to reduce language to a purely temporal code, as it is only defined by the duration and the order of the intervals, allowed to apply a challenging task which at the same moment ensured that any learning effect would be a “pure” product of the temporal training per se. The potential effects of this learning procedure would be assessed through multiple temporal tasks measuring temporal order discrimination (i.e., a temporal order judgment task) and duration estimation (i.e., a temporal bisection task) allowing to examine for a global change in temporal perception.
The experiments conducted in Chapter 4 involved unisensory (i.e., visual and auditory) and multisensory (i.e., audiovisual) Morse code learning versions as a part of an eight-day training procedure. The analysis for the three sets of experiments reported in Chapter 4 demonstrated, for the first time, that when people were trained with an engaging and challenging methodological approach as the Morse code learning task, training-induced changes could be observed, translated into improved temporal perception skills. Specifically, the improved ability to detect temporal asynchronies (i.e., significant narrowing of the temporal window along with a reduction in the just noticeable difference) observed for the multisensory group for both auditory and visual order judgments indicated a higher learning and generalization effect among untrained modalities as compared to the unisensory training. In the same wavelength, the multisensory benefits of learning were also apparent for the temporal bisection task, where a sharpened temporal estimation (i.e., the point of subjective equality approached the real bisection point) and an increased temporal sensitivity (i.e., a significant reduction in the just noticeable difference) after training, reflected a “solid” indication for a real-time perception change and temporal plasticity.
The robust temporal perceptual learning effect that was attributed to the multisensory Morse code learning task (Chapter 4) was further investigated on whether it could potentially last through a long-term period without any daily temporal training (Chapter 5). The results indicated that the observed multisensory training-induced changes were stable and could be maintained at least over a period of one month from the final day of training, with the narrowing of the temporal windows for the order judgments and the reduced just noticeable differences for the duration estimations remaining in similar -to the post-learning phase- levels for the follow-up assessment. The one-month gap away from the learning task, except that it was in favor of displaying the maintenance of learning, also helped to exclude the possibility that the observed learning effects could be due to the repeated training exposure.
A further evaluation of the Morse code methodological approach was attempted in Chapter 6, where the aim was to optimize the efficiency of temporal perceptual training by adopting alternative learning strategies (i.e., active learning and a temporally enriched environment). This was achieved by asking the participants not to just recognize the alphabet letters, but to reproduce the Morse code patterns through a temporal reproduction task. The comparison between the learning tasks of Chapters 4 and 6 showed non-significant differences in the temporal sensitivity to their order judgments and the duration estimations, supporting that the temporal benefits gained from the two different learning tasks were similar. Thus, the hypothesis that the temporal perceptual learning effects would be higher for the more explicit temporal environment and the active response design followed in Chapter 6, was not obtained, with the lack of informative feedback during the multisensory training, being considered as the main responsible for this inability.
Overall, the findings of the present thesis provided the first empirical evidence that a more engaging and temporally related multisensory methodology was able to improve both temporal discrimination and estimation skills across modalities. Specific disorders (namely schizophrenia, autism spectrum disorder, and dementia) have been found to be linked with impaired temporal order or duration deficits (e.g., Vatakis & Allman, 2015), or else, the aspects of temporal perception that found to be improved in healthy populations through the novel learning approach followed in the current thesis. Thus, our learning scheme could also potentially contribute as an effective diagnostic and interventional strategy for cases of non-invasive, behavioral, temporal training, in an attempt to prevent, delay, or ameliorate potential timing distortions in clinical populations.
