These workshops and short courses are organised by the CogNovo Research Fellows and their supervisors, on an ad-hoc basis, for the sharing of skills, knowledge and research methods.
One example is the series of WHAT’s UP - a weekly time slot to present new ideas, technical, and methodological background or start a discussion about some interesting topic. WHAT's UP is curated by Yaniv and Frank. Please drop them a line if you want to present something in that series.
Whilst primarily of interest to our CogNovo Fellows, all of these events are open for all to attend.
Unless otherwise indicated, they all take place in the CogNovo Headquarters, Link 3 Seminar Room and Workspace, Plymouth University
…is intended to start a discussion about how we could easily communicate and will shed light on possible uses of some applications installed on the computers provided by the university
The idea is that we address things on a strictly introductory level, suitable for everybody. You bring questions, comments, data along and we work through them as far as possible as a group, on laptops, in the big space.
If nobody brings anything I will pick a small topic and run through a few tips and tricks.
Roman Borisyuk is going to discuss the Friday-Puzzles and ask your opinion if these puzzles are useful, interesting, difficult, easy, etc. Also, he would like to know how many students were able to solve them. Also, he will discuss solutions and around.
Creativity, according to the creative genius Leonardo Da Vinci, is doing something else when others have copied what you’ve done before. Creativity to me is a cognitive property which allows us to be motivated to generate something which has never existed before. I am sure one could come up with a simpler or more complex definition of creativity. Definitions of creativity can be extended beyond the basic belief of it being related to something new. It can include concepts like (i) creating using what, (ii) in what ways, (iii) using what methods, and (iv) how widely is the creation accepted by others.
Scientific fields like Psychology and Cognitive Neuroscience have been trying to understand this elusive psychological construct using a wide variety of definitions, understandings and assumptions. We as a part of CogNovo, a programme aimed at understanding creativity, will soon start using our independent definitions of creativity. These definitions will be guided by our motivations, ease of use, availability of behavioural tests and a wide variety of other factors. Let’s come together at this week’s WHAT’s UP and discuss what we think creativity is, and why we think our definition is more suitable. Additionally, and more importantly if possible, come at a common definition of this construct we all aim to chase.
With creativity being such an important factor for many of the projects in CogNovo, we thought it would be useful to join forces and create a creativity database. This database will allow (anyone who is interested) to record what they have read about creativity as a process, person, place etc. More importantly, everyone else will then be able to view this record and access the paper.
We will be demonstrating how this database works during the What's Up session. In the meantime, take a look at the database and prepare any questions or additional sections you might want to add to it.
If you have technical issues concerning access rights to the database let Frank know.
Working with colleagues in Zurich and Bielefeld we have published two papers in the last 18 months that are the first in the literature to propose a neuromorphic approach to devices that model adaptive auditory feature extraction. We are now exploiting this approach, and exploring other approaches, to work towards a device that could handle real sounds and learn to recognise "auditory objects" in field conditions.
Some might argue this is not a topic to talk about with primary or even high school students. Being a STEM volunteer I do exactly that and believe it is a jumping off point to act as an "ambassador" for science and engineering in general. This includes mathematics, computing, the environment, the Cognition Institute, the University and so on in all their infinite variety.
Basic topics including:
What is Prezi? Getting started with Prezi; Add text; Choose style; Create lines and frames; Add images; Use of video in Prezi; Make a path in your prezi; Download Prezi.
Moving on to more advanced topic such as:
3D backgrounds; Create gradients in the background; The difference between vectors and pixels; How do create infographics in Prezi (this topic will be covered quite simplistically as it required knowledge in different design imagine software); Fade-in animations; Adjust Frame thickness; Images in Prezi; The use of sound/voice over; Prezi embed in a website; Plus others
Better structuring in Prezi; Better building your prezi; Thinking of a good strategy; Creating a clear overview; How do you get creative?; How do you tell your story?; Surprise effects
This WHAT’s UP session will focus on Praat, a widely used open source software in the world of phonetics and speech science. Apart from being an excellent tool to carry out acoustic analyses and some basic speech synthesis it is an easy way to work on sound files in .wav or .mp3 formats. For example, it is possible to add silence of any specified duration to a sound file, to concatenate different sound files, to cut and paste portions of the sound files, to add labels etc. At a more advanced level, it also allows the use of scripts, which can save a lot of time when performing many sequences of operations. Praat was developed by Paul Boersma and David Weenink in Amsterdam in 1995, and it has been constantly updated ever since (download and description here).
The session will be an overview of the many possibilities that Praat offers to edit sounds and analyse them. In case there is an interest in learning how to perform some more specific operations, I will prepare other sessions to cover more advanced topics.
Pure Data (aka Pd) is an open source visual programming language. Pd enables musicians, visual artists, performers, researchers, and developers to create software graphically, without writing lines of code. Pd is used to process and generate sound, video, 2D/3D graphics, and interface sensors, input devices, and MIDI. Pd can easily work over local and remote networks to integrate wearable technology, motor systems, lighting rigs, and other equipment. Pd is suitable for learning basic multimedia processing and visual programming methods as well as for realizing complex systems for large-scale projects.” (from: https://puredata.info/)
In this 1 hour workshop I will give you an introduction to Pd and show what others have done with it. Focus will be on practical exercises - you will write little programs, called “patches" in Pd. Who those who are interested we can do some in-depth stuff in a later session.
We will discuss why Python is one of the most popular languages nowadays, increasingly replacing Matlab, R and some other programming languages. We will see the domains where it is used, with a particular focus on using Python for scientific purposes. Whether you do data analysis, language processing, image processing, computational modelling, experimental design or statistical analysis, Python can help you with that. We will present many of the Python packages that are useful in linguistics, psychology as well as in computing areas.
Docear is yet another software to manage your citations. The developers call it an academic literature suite. For us it looks like a mixture between a mind mapping, citation tool, and an idea scratch book. In that sense it might be helpful to get the read literature organised.
We will show you how we use it and might gather some ideas how to improve that process.
This will be a hands on session – so bring your devices! …and we hope rather to start a discussion than giving a talk…
This brief and informal workshop introduces Zotero, an open source citation management tool. Troubleshooting questions specific to Zotero will be welcome. However, the workshop will also use Zotero as a mere starting point for further discussion on strategies for (collaborative) bibliography management within CogNovo. Comparisons between citation management tools will be encouraged.
They will present their ideas on how to proceed in their research a.k.a. RDC.1 talks. Please expect interesting insights on neural concept samplers, computational brain modelling, engaging and trustworthy voices, and mixed reality games
In a genetic algorithm, a population of candidate solutions (called individuals, creatures, or phenotypes) to an optimization problem is evolved toward better solutions. Each candidate solution has a set of properties (its chromosomes or genotype) which can be mutated and altered. Each generation, which are basically solutions generated at the same time, is tested against a fitness function for it's validity to solve the problem. The fittest solutions will be combined to new generations and tested iteratively.
Jack will give a general introduction on what those algorithms can do for you and will discuss how they could potentially be used to solve your problems.
In this talk, John will be introducing a number of the basic technical aspects of photography and cameras. It will cover Exposure, Depth of field, Shutter speed & Aperture, Light & Colour Temperature, ISO settings, focusing modes, Focal lengths, Image Quality settings. It’s suitable for anyone interested in better understanding both consumer model and SLR camera operation, though this talk will have minimal application to iPhone cameras.
The remote control of mechanical devices equipped with a large number of actuators, such as humanoid robots, is a challenging task. When dealing with the resulting large number of degrees-of-freedom, the nature of the interface provided to the human operator plays a fundamental role in the success of tele-robotic performance. A wide range of tele-robotic interfaces have been explored so far; some are very rigid devices that require a great deal of cognitive and manual effort, while other more intuitive systems, based on one-to-one body mapping, are in contrast very complex and expensive devices, often specifically tailored to a single robotic platform. In this What's up talk Christos will describe his approach to interfacing with robots.
Reinforcement learning (RL) is likely to be at the very heart of how we learn, act, create. It has a long multidisciplinary history spanning all of cognitive science. Due to this, the words "reinforcement learning" have different meanings in different contexts, and misunderstandings are common across disciplines. What does RL refer to in psychology, artificial intelligence, neuroscience? What are its most basic mechanisms? What are the latest developments – and what do they mean for your discipline?
Thomas will try to answer all of those questions - and discuss any other question you might have.
GIT and github are great tools to manage and share data that changes over time. While GIT was invented to track changes of thousands of developers in one of the largest shared code bases it's basic functionality is easy to use. Frank will discuss how these basic functions can be used in setting up experiments, analysing data, and getting feedback and help from other CogNovians for these artifacts. If you are undecided if or how to use it - this talk will give you a basic understanding and you will be able to decide afterwards if it is something you would like to use.
The question of whether switching patterns in visual and auditory multistable stimuli are domain specific or domain general is open. Some studies show that auditory and visual ambiguous switching patterns show similar characteristics (Pressnitzer & Hupé , 2006) while other studies point out important differences in the auditory and visual phenomena (Kashino & Kondo, 2012). The current aim was to address the question of domain generality/specificity in the number of perceptual switches and the phase duration of the initial interpretation in both visual and auditory bistable stimuli. Also, we examined whether inhibitory capacity, working memory and set-shifting abilities relate with the number of switches and the phase duration of the initial interpretation in both types of bistable stimuli. We conducted two experiments. In both experiment we measured participants’ switching rates across 5 trials. In the first experiment (N = 20) we used the duck/rabbit ambiguous figure (Ambiguous figure task) and the fly/life verbal transformation stimulus (Verbal transformation task). In the second experiment (N = 19), we used the cylinder moving clockwise/anti-clockwise (Ambiguous motion task) and 1tap/2 taps (Auditory streaming task). Further, participants completed an inhibition task (Day Night Stroop task), a working memory control task (Day Night Memory Control Task) and a cognitive flexibility task (Fist). We found a high correlation between the numbers of switches in the duck/rabbit ambiguous figures and verbal transformation task. This may be preliminary evidence of domain general switching patterns in high-level auditory and visual stimuli. In contrast, no such association was found for ambiguous motion stimuli and auditory streaming stimuli. Thus, there is no evidence of general switching patterns across domains when low-level auditory and visual stimuli are used. Inhibition, working memory and set-shifting did not relate to switching patterns within and/or across domains.
Guy will present a 15 minute talk which will detail different phases in the production of a 16mm colour sound film which he finished this year. He will give some insight into the techniques employed and also reflect on the process of working with technologies which are on the brink of obsolescence. Following the talk he will screen the 16mm print of the film (Bugatti - Le Terrain Anglais, colour, optical sound, 15mins) and welcomes questions after the screening.
In preparation for an EEG workshop that is going to happen next week, Frank is moderating an installfest for required software. Please bring a bit of patience and your computers - if possible with a working MATLAB installation. We will start around 4pm. If you can't make that - just join in any time later or contact Frank if you need help. After the session everyone should have MATLAB, EEG Lab, and ERP Lab running. Also, if you want to collect data yourself, we will try to install OpenVIBE.
The software is a prerequisite for participating in the workshop - so please make sure that you will attend today's session if you signed up for the workshop next week. You are also welcome if you do not participate in the workshop but need support to set up this software.
In previous research Diego had done on dance styles from underground and urban subcultures, he argued that technology is changing the way humans dance and predicted that we will continue to see new technological metaphors translated into expressive movement vocabularies. For Cognovo Project 8, he aims to extend this argument by exploring how the Internet is changing patterns of artistic creativity. Focusing on music as the central creative activity, he will speculate on how the Internet will (if at all it does) influence creative cognition and behaviour in the near and far future. Diego will outline his current research strategy in this talk in order to invite feedback from the audience.
The work of Darya Zabelina and Mark Beeman refines the long-standing contention that "creative people" are generally more distractible or have leaky attention. They demonstrate that different types of creativity emphasize different types of attention. Specifically, people with real-world creative achievements may have leaky attention filters, while people who perform well on timed laboratory tests of divergent thinking (which are often used as a proxy for creative cognition) appear to have selective, yet flexible attention. Their behavioral and EEG experiments suggest that real-world creative achievers make more errors when switching levels of attention, show more interference from uncued levels of attention, and show less gating of sensory information as assessed by the P50 ERP, suggesting leaky attention. Conversely, divergent thinkers exhibit attentional flexibility on a selective attention task, and show more gating of sensory information, as assessed by the P50 ERP, suggesting selective attention. In addition to speaking about creative cognition research, Darya will also share her experience as a PhD student in Cognitive Neuroscience at Northwestern University.
Tara will give a brief and informal introduction into the design and findings of her second experiment, which was investigating blocking and its relationship with schizotypy and creativity.
In this presentation, we explore different surviving versions and varying restoration approaches regarding one of the most canonical avant-garde film of the 1920s, Ballet Mécanique (1924). A defining work of avant-garde film, it is known to many people who have not even encountered it in its full form, rather just from photographs, as part of the oeuvre of Fernand Leger, or as part of the musical works of George Antheil. Those who have seen the film may know it from any different number of versions, most - but not all of them – in black and white. Based on a unique colour print surviving in the collections of EYE Film Institute, a more recent restoration was carried out in colour, and concepts and studies towards potential hand-applied, dye-chemical replication of its original hand-colouring were concurrently developed. In outlining the history of the film's multiple restorations, we highlight past developments, current approaches and future opportunities in colour restoration.
Though CogNovo aims to be a leader in interdisciplinary research in creativity, the question remains open on how to practically implement creative solutions to everyday challenges. The challenge that we would like to talk about, and that is an important topic for us, is that of sustainability. How can we as individuals make a difference to the environment using creative strategies and through introducing new habits in our daily routines? We will share some of our practices with you, with the hope of breaking ground for longer lasting discussions on the topic.
Regional accents may affect initial judgements of personality traits, such as trustworthiness. We examined whether accents interact with behavioural patterns to modify initial trustworthiness attributions, using an iterated trust game in which participants make investments with virtual players, which are associated with two British English accents.
Guy will recount details of his recent trip to Indonesia as film preservation special envoy and guest of the Europe on Screen film festival. Aside from contributing to public Q&A sessions, Guy also visited the Sinematek Indonesia where he exchanged knowledge with staff and experienced at first hand the challenging conditions of film archiving in the tropics.
The Genode OS Framework is a tool kit for building highly secure special-purpose operating systems. It scales from embedded systems with as little as 4 MB of memory to highly dynamic general-purpose workloads.
Genode Labs, the company behind Genode OS, has spun off Technical University of Dresden a few years ago. Sebastian, who has been part of this start up since the beginning, can provide some insight on technical details as well as an overview on how a company can exists that is purely producing open source software. The whole session will be rather a chat than an actual presentation.
I will discuss interdisciplinarity at the conference Reinforcement Learning and Decision Making 2015 (RLDM 2015 for short), which I attended earlier this month. RLDM brings together the researchers interested in learning to act from experience. These researchers come from a variety of fields, including artificial intelligence, computational neuroscience, and cognitive psychology; their backgrounds and research methods vary considerably. I will give a rough interdisciplinary introduction to reinforcement learning. Then I will move on to a discussion of the difficulties of organizing an interdisciplinary conference. Can communication between scientific fields be productive? In what ways? What are the pitfalls of the exercise, important solutions, and potential improvements? This talk may be useful to those of us planning to organize or participate in interdisciplinary conferences - e.g. Off The Lip.
After RLDM, I took the chance to visit the Canadian Rocky Mountains around Banff and, stopping-over in Iceland on the way back, Reykjavik and the Snæfellsnes peninsula (for those not familiar with the area, it is located west of Borgarfjörður). I'll show a few pictures to those interested while everyone eats cake.
Konnichiwa! In this talk Ivana will first briefly talk about the summer school she attended in Okinawa and then she will take you on a journey through a country full of astonishing contrasts. The summer school in computational neuroscience is an annual event organised by the Okinawa Institute for Science and Technology (OIST) and it gathers scientists in experimental and theoretical neuroscience. Ivana will explain why this is so important if we want to understand how our brains work, and tell you about what she learned and brought back to her PhD.
Then, she will tell you about her two week travels across the mainland Japan. We will start the journey in Hiroshima and hop along the cities on the eastern coast to end up in Tokyo, which she explored with another Cognovian. The story includes incredibly friendly people, big bright cities, amazing (and less amazing) food, random welcoming strangers, beautiful nature and high-tech toirets.
Despite her best efforts she did not manage to eat all the sweets she brought from Japan, so there will be some matcha daifuku mochi (first come first served!) and green tea (if we manage to decipher the preparation instructions).
How are we able to learn associations between events and outcomes? What determines how fast we will learn these associations? How will this change throughout the course of learning? Associative learning models have attempted to answer these questions. This talk will give a brief introduction to some of the most influential models of associative learning from a Psychology perspective.