- Consciousness is a much debated yet less known topic that has been debated by both science and philosophy. However most of our knowledge about consciousness stems from introspection and behaviour analysis. This paper helps us move away from the psychological intuitions of consciousness.
- Through well-designed examples, the author allows us to understand that we cannot be the judge of our own consciousness.
- We may think that we see the full picture but our consciousness consists of a small fraction of what is actually in front of us. This theory is well explained through the “ Whole report paradigm”
- After providing a convincing amount of examples about why the human iconic memory is a fragile form of working memory, Lamme moves on to showcase that the mind is grossly incapable of comprehending itself. Time and againLamme exposes the cognitivegap forbids us from understanding the true nature of consciousness.
- Lamme frequently refers to the “Global Workspace Theory” to help readers understand that human consciousness is never about seeing the bigger picture but about putting seemingly important things under the spotlight and keeping others in the background.
- Throughout the paper,Lamme stresses that seeing is rich but our iconic memory falters when it comes to reporting what we saw.
An ecological valence theory of human color preference
Ecological valence Theory proposes that people like colours in the same degree as they like objects associated with those colours. The ecological valence theory also predicts that the preference decreases with the increase in the number of associations a colorhas with objects. Palmer shows that Ecological valence theory is a plausible explanation to the field of human color preference.
Although color preference drives a lot of real world decision-making in humans, there still is very sparse literature on why we like certain colors and dislike others.Hurlbert and Ling’s theory that color vision developed as a evolutionary strategy so that females could differentiate between
The core of the Ecological valence theory dictates that the average color preference of people are determined by the average affective valence of their responses to objects that are usually associated with that particular color. A study was conducted where participants were asked to rate the 32 chromatic colors of the Berkeleycolor project. The Weighted Affective Valence Estimate (WAVE) for each of the 32 colors were calculated and the results were corresponded to the color preferences shown by the subjects. Subsequently other participants were shown the object descriptions from the original study group.
The study found that the average color preference of the sampled individuals ( Berkeley, CA) was strongly correlated with the object preferences of an independent but similar sample of people. However, Palmer fails to shed light on the origins of these strong color associations; whether they are hardwired or learned color preferences. The data also shows that sometimes social investment can influence people’s color preference to a great degree. One confounding factor in the study that was pointed by Palmer was that people could be listing out more desirable objects for the colors they like and list less desirable objects for the colors they dislike.
Eye movements in Language and Cognition
Daniel C. Richardson, Rick Dale, Michael J. Spivey
As the authors correctly point out, eye movements have low metabolic costs and therefore their activation thresholds are lower. The ‘Ceaseless twitching’ helps human beings process the large amount of visual information constantly thrown at us by our surroundings. The authors draw our attention to the Saccade eye movement, which is due to eye’s attempt to focus objects on the fovea. Experimental psychologists usually like to observe the pattern of locations of the saccades called the scan path. The duration between the stimulus and the saccade is called the saccade latency; the time of focus is known as fixation duration.
In this interesting chapter, the authors try to provide the readers with the reasons why eye movement tracking could be a window to offline cognitive processes. Research has shown that the eye movement pattern shown by subjects during offline cognitive functions often mimic the online cognitive processes such as object manipulation.
The chapter also focuses on the eyes extraordinary ability to use semantic information for spatial indexing has been well illustrated through examples.Recent research has revealed that eye movement technology can be used to understand how listeners are hypersensitive to nuances of language. Nuances of language are capable of driving eye movements and eye movements on the other hand can be effectively used to understand the incremental and interactive nature of spoken language comprehension.
Despite the advantages of monitoring eye movement, the methodology is faced with same constraints. The primary constraint is that eye movement cannot be observed under natural conditions. Researchers strongly believe that the eye movements are not the same in sterile laboratory conditions. Despite limitations, eye movement research has made a lot of advances in the past few years. By furthering the prospect of eye movement study we will gain better insights into cognitive linguistics and offer better explanation about how cognition happens.
Generating with Recurrent Neural Networks
Recurrent Neural Networks presently do not enjoy the widespread use due to their apparent untrainability. However with Hessian-free optimizations, the true strength of the Recurrent Neural Networks ( RNNs) can be finally unleashed. The paper talks about the ability of RNNs to perform character level-language modelling tasks. The paper talks about the deficiencies of the present RNN architecture and how that could beone of the contributing factors to its untrainability. The hessian free optimizers on the other hand can be used to train the RNN to a level that was previously unknown.
Language modelling is very difficult because constructing sentences is not about creating the right array of word, but also about having a message behind it. Therefore, in order to make language modelling to improve their understanding of the text. With the advancement of character-level language modelling people will be able to interact with computers in a better fashion. Ideally RNN is a very powerful tool as its weights are shared across time and it also features no-linear activation function that are used by its hidden units.
Most of the experiment results show that the MRNNs are exceptionally adept at learning words. Additionally, MRNNs were found to deal exceptionally well with words that did not appear in the training set. By only learning 1500 hidden units, MRNNs have outperformed the memoizer. I agree with the authors claim that if MRNNs are trained with millions of units of connections they can achieve a higher standard of performance albeit at the cost of computing power.
Cognitive Nueroscience: Targeting Neuroplasticity with Neural Decoding and Biofeedback
All neuroscientists want to answer the fundamental question of howsignals from the brain turn into behaviour. Establishing causality between brain and motor behaviour functions. The paper gives us a delightful insight into how neuroscience can help us understand individual brain region functions. The two approaches discussed in the paper providea way in which the decoded neural signals can be used as biofeedback to induce targeted neural plasticity.
For a large portion of the last five decades, scientists have tried to unearth a brain map that designates areas of brain that are important for certain cognitive functions. The methods discussed in this paper are path breaking as they move away from the beaten track. Instead of making participants perform a task, they are instructed to alter brain activity through biofeedback, this can lead to elegant insights about which region of brain is responsible for what cognitive function.
The innovative use of FMRI technology in this paper is truly one of the biggest contributor to the astounding results. In the earlier days, participants would be asked to perform certain tasks after a certain amount of training and then the brain scan would show activated regions. However, the inherent flaw in this method lies in the fact that the training and execution of certain tasks require contributions from different areas of the brain. FMRI and other improved brain imaging techniques have allowed us to understand not only the role of each brain region in learning but also the plasticity mechanisms that occur in these areas.
Another unique fact about this study was that participants did not have to go through a rigorous training and practice process and were unbeknownst to the stimuli that the investigators were testing.
A very insightful work on how cognitive neuroscience can prosper by targeting neuroplasticity with neural decoding.