1.1 Definition: Simulation theory (ST) in contrast to TT dismisses the idea that we use a specific ‘theory’ to understand other people’s minds but rather we mentally simulate them. According to ST, to understand others, we utilize our own mental resources and ask the question “what would I have done if I were in his position” or, simply put we “put ourselves in their shoes”. However, as argued by Gordon (1986, 1992), simulating other minds is not merely projecting our own situation but also accompanying it with ‘necessary adjustments’. Thereby, the ST account is contrary to having a generalized mechanical law or a theory about the individuals around us. The ST has also been characterized as information poor mind reading (Goldman, 1992) as it does not require a huge set of information or theories as required by TT. 1.2 An example: Consider the case where you are playing a football match and see your opponent pass the ball in the opposite direction of the goal and you search for a possible explanation as to why he might have acted in this way. While searching you find that from the perspective of your opponent, there were none of his team members near the goal. Upon simulating what you would have done in such a case you arrive at the explanation that it would have been safer to pass the ball in the opposite direction which matched with the performed act. In this example, you use your own mental resources to comprehend other’s intention based on the assumption that they would be similar to yours. It is also clear that this simulation required not just projecting oneself in another's situation but also making the necessary adjustments while taking into account the deviating goals and motivation of the other individual. This explanation is contrary to the one that would be provided by a theory-theorists. A theory-theorists would explain this phenomenon by referring to the general assumption that ‘when players don’t find same team player next to them in the forward direction, they pass the ball to another same team player in the opposite direction’. 1.3 Evidences for simulation theory: ST has been suggested to be routedin basic old and simple mechanisms, taking advantage, for instance, of human abilities to read gaze direction or to imitate others (Gallese & Goldman, 1998; Gordon, 1992). Proponents of ST often cite examples of egocentric errors or biases – i.e influences of the mindreader’s own mental states such as beliefs or desires in understanding others (Goldman & Jordan, 2013) - to provide evidence for ST. ST has also found support from what is known as the ‘curse of knowledge” – a state where even though the participants are informed of the difference between their own knowledge and target’s knowledge, participant’s own knowledge seems to influence target’s attributes (Birch & Bloom, 2003; Camerer, Loewenstein, & Weber, 1989; Nickerson, 1999). ST has also gathered support from neurophysiological studies where, for instance, amygdala lesion in patients reduced their recognition of emotion in facial expressions (Adolphs, Tranel, Damasio, & Damasio, 1994) as well as discovery of mirror neurons proposed to be a neural implementation of simulation (Gallese & Goldman, 1998; Rizzolatti & Craighero, 2004). 2.1 Formulation: ST has been difficult to assess because of the confusion in its conceptualization. There has been questions as to what could be termed as simulation. Different proponents of the theory have slightly different interpretations as to what would count as a simulation. Over the years, ST has been refined and reformulated. Goldman (2006) introduces simulation as a process in the following manner: “Process P simulates process P´ iff (1) P duplicates, replicates or resembles P´ in some significant respect and (2) in its duplication of P´, P fulfills one of its purposes or functions. For mindreading simulation, P and P´ are mental processes and the purpose or function of P – the simulating process – is to understand a target’s mental states”. 2.2 Abstract or concrete: This definition obviously raises the questions as to what would constitute as “some significant respect”. It has been argued that it is not sufficient for simulation to merely duplicate another process but it must involve a concretely similar processes (Fisher, 2006). A concrete replication would imply a similar set of systems or fine grained similarities in the replication. In contrast, an abstract replication might involve a different set of processes to replicate the given process. Imagine for instance the case where, you would need to recognize a given human face. A human would utilize a specific set of processes while a computer uses a completely different set of rules to identify the given face. In this case, the computer is simulating the process in an abstract manner. (Fisher, 2006) and (Spaulding, 2010) claim that simulation as referred by Goldman and others has to simulate concretely to claim non-trivial effects. Understanding this criteria is also important to clarify genuine cases of simulation compared to cases where a process just ‘happens’ to get the correct conclusion. Consider the case where your friend receives a gift certificate for a supermarket and you theorize that since he desperately needs a coffee machine and he is an Italian who really enjoys coffee, he would buy a coffee machine for himself. Suppose, in this case, your reasoning just so happens to be the same as your friend’s (and he uses same reasons and buys a coffee machine). Despite the fact that you theorized, your reasoning process accidently resembled your friend’s reasoning process. This similarity of processes has been referred to as an abstract similarity or abstract replication (Fisher, 2006). It is clear from the example that abstract replication would be entailed by at least some versions of TT. 2.3: Fulfilling function: It is also not clear when a process fulfills the function of another process? For a genuine simulation, one must distinguish between cases where a process happens to replicate another versus cases where the function of the process is to replicate. For example, in understanding where a tennis player will serve, I may act as a ‘fact-finder’ and theorize it based on the position where his opponent is standing. In this case, I use my visual system not for a simulation but for deriving a theory. So, even though, I successfully predict the outcome, my process merely happens to replicate the process going on in the target and thus would not be a simulation. 2.4: Strict formulation of simulation: In view of these points, it’s imperative to develop a stricter formulation of what constitutes a simulation. The simulation, thus, needs to concretely replicate the target’s process, i.e. have fine-grained similarities. The simulation also needs to be ‘offline’ utilizing non-standard inputs, i.e. it should utilize the same processes as in the target but without providing a final output in terms of an action. In addition, the inputs and outputs of the simulated procedure should also be the same as that of the target. 3.1 Classification : Having defined what a simulation needs to be, simulation could be classified on basis of:
3.2 Retrodictive and Projective simulations: Retrodictive simulation concerns the cases where explanations are generated after the behavior has been observed. Such simulations help us understand why an actor performed an action retrospectively (Gallese & Goldman, 1998; Goldman, 1989; Spaulding, 2010). Projective simulations, on the other hand are used to predict the consequences of others’ actions. This takes in retrodictive simulation as an input. The aim here is to predict what an individual will do next (Gordon, 1992; Spaulding, 2010). 3.3 High level or low level simulation: A high level simulation concerns mental states of complex nature such as beliefs and propositional attitudes and has been defined to have at least some degree of accessibility to consciousness and voluntary control (Goldman, 2006). Earlier versions of ST were thought to be comprised only of such high level simulations. A high level simulation would consist of imagination in the usual sense where one consciously puts oneself in someone else’s position and imagine what he/she would feel or do in such a scenario. A low level simulation on the other hand, does not have conscious access or control over the simulation process. The simulation has been proposed to work unconsciously and automatically (Gordon, 1992). A neural realization of low level simulation has been proposed to be accomplished by an action observation/execution matching system ( mirror neuron system) (Gallese & Goldman, 1998). 3.3.1 Mirror neurons: This class of neurons, first discovered in monkeys, is active both while the individual performs an action and when the same individual observes another individual perform the same action (di Pellegrino, Fadiga, Fogassi, Gallese, & Rizzolatti, 1992; Rizzolatti & Craighero, 2004). It has been proposed that when we observe someone perform an action, activation in our mirror neuron system activates our brain ‘as if’ we were performing the same act. An inverse model that translates from observed actions to the respective motor commands is thought to play a role in determining the specific motor commands that may have caused that action (Kilner, Friston, & Frith, 2007). Mirror neuron system, thus, has been proposed to be the basis for our understanding of others (Keysers & Gazzola, 2009; Rizzolatti & Craighero, 2004). 4.1 Limitations: Despite the supporting evidences for ST, there are many issues that face the theory: 4.1.1 Computational cost: This limitation primarily concerns high level simulation where one needs to track the possible mental states that caused the behavior. This can get relatively complex and computationally challenging given the complex social interactions involving multiple individuals and their interdependent behaviors. 4.1.2 Threat of collapse: Both low level and high level simulations for retrodictive case would loop infinitely and would have no stopping point as there is no way to confidently determine if we have arrived at the best explanation (Davies & Stone, 1995) 4.2 Inapplicable cases: One can think of many cases where ST can’t be applied. For instance, in cases of delusion of persecution or individuals who we know to exhibit idiosyncratic, irrational behavior, ST is irrelevant since our own experiences are too different from such individuals (Newen, 2015; Tversky & Kahneman, 1974). 4.3 Argument from systematic errors: ST has also been criticized for not being able to explain the errors exhibited by children and adults (Saxe, 2005; Stich and Nichols 1995). For instance, egocentric errors in children for certain mental states are corrected later than others. It has been proven for instance that the mental ability to understand the perceptions and desires of other individuals occurs earlier in infantile development than the ability to ascribe belief which misrepresent the reality from the own point of view.(Gopnik & Wellman, 1992). 4.4 Self vs other distinction: An “offline” simulation requires similar substrates as in the target would invariantly require the distinction that the action is performed by the other individual and not by the individual who is simulating. This distinction would require at least certain non-overlapping brain regions (Becchio, Adenzato, & Bara, 2006). 5. Conclusions: Some of the criticisms of ST have led to the refinement of the theory. Strong versions of ST (pure simulation) have largely been dropped in favor of hybrid ST/TT models (Davies & Stone, 1995; Goldman, 2006). Hybrid models with an input from theoretical knowledge from TT can solve the limitations of computational cost as well as threat of collapse. Hybrid models where simulation might not play an exhaustive role and proposing two or more cognitive processes driving simulation would explain different developmental time spans for beliefs vs desires (Goldman & Sebanz, 2005; Goldman, 2006). Despite these reformulations, ST and TT remain limited in the sense that they adopt a third-person perspective where the simulating individual is an observer (Gallagher, 2008). Our requirements for understanding others’ actions, however, is critical while we are interacting with them rather than being in a passive retrospective condition. Both ST and TT accounts thus fail in explaining such situations. References:
0 Comments
Disclaimer: This is not supposed to be a bashing of a specific reviewing system or any specific person but a frustration of the reviewing system in general. I might yet be in the 4th stage of grief.
Rejections are almost always bad (except probably some) and I guess all researchers have faced them at some point - be it a grant application or a paper. So, rejections are not really new to us (See @YourPaperSucks and many others). But out of all those harsh comments, there are some that prick the most. Sometimes for a good reason - a simpler viewpoint of the data that fits better with the data but not with your idea or a control experiment that you should have performed or a bad statistical error. But some others hurt not because of the soundness of the paper but because of prejudices and opposite viewpoints. This bad rejection is about the latter. As a context, the review that I refer to, was not from a journal but from a very interesting conference. Of course, I would be biased in evaluating the interest level and the usefulness of my personal work and it would be understandable if the conference editors decide that the work was not interesting enough for them. However, I take exception to two specific points:
While surely, I am not the first one that has commented on these topics and twitter especially is full of lengthy discussions on these topics, my point is that it's high time that we start taking these concerns seriously. The tone of the review. No matter how badly a paper might be written or a study performed, no one should undermine others work as "The authors perform a relatively simple study " or "Useful experimental fact, but not totally unexpected that this can be made work,". Each study designed not only requires a huge amount of time to execute but also weeks (and often months) of planning. So, in my opinion, to say that someone performed a simple study is to disregard all the hard work authors put in the study. Again, we are all aware of the Hindsight bias or knew-it-all-along effect. So to say that a work is not totally unexpected, in my opinion, is to lower the efforts of the authors (not to mention the huge importance of replicating multiple effects). Prejudices against some theories. While I agree that we all have our pet theories and it's probably perfectly fine to be biased to them (as long as you are open to alternate views given data), it is not proper to be disrespectful to views of the others. I believe, propositions of the order "skip grand conclusions." are misplaced in a scientific review. Overall, while my blog might have stemmed from the anguish and frustration of the review, I am positive that the more everyone is aware of such issues, the better it should get. A similar (more positive) position has been voiced by a recent nature news article that talks about the prejudices in research and does a nice job at summarizing the issues. Three months ago, I was thrown into the Italian world when I joined the lab there. The experience has been awesome, stupendous. I have been blown away by the beauty of Italy. Let me tell you that there is a lot of difference between the photos that you see on the internet and the actual places. The pics are, I think, one dimensional.
However, I am not going to talk about something that a lot of people have already talked about and a lot better than what I possibly can. I am about to write about the not so great aspect of coming to Italy for people who don't know Italian. As you already know, I am an Indian and so have absolutely no idea of Italian language. It was difficult for me and even though I did start practicing the language like 2 months before arriving, I felt it was not enough. There were certain things that I wished I knew before coming that would have really helped me a lot. However, I believe these things might help you. What I am going to write down are certain sentences/phrases that will help you in the initial period here. This is in no way a substitution for learning Italian and certainly no excuse for not learning the language. I strongly urge you to learn it once you are here. I will also provide you a list of places in Genoa where you can learn Italian. Ok, so let's start. I will start with some distinctions between English and Italian and some suggestions which probably help understand what to use and importantly what not to!! 1. What is important to know is that even though English does not have distinction between formal and informal words, it is widely prevalent in Italian (and other languages). So, it might sound rude if you talk in an informal way to a stranger elder to you. (although i have never found Italians to be angry if you use the language improperly but instead they offer you help). 2. The construction of sentences is very similar to English and a lot of the words are actually very similar to English. Probably - probabilmente. type - tipo. case - caso. Of course, the pronunciation is different. 3. While learning Italian, don't think you are learning English and translate it directly (although it's unavoidable) but try and understand that the way some things are said (and thought) in Italian are not the same as in English. For example, although the literal translation of "what is your name?" in Italian would be "qual è il tuo nome", this is not the way it is said in Italian. Instead, "come ti chiami" is used which literaly translated to English would be "How do you call yourself?. So the point is that don't memorize that what's your name in Italian is "come ti chiami" but instead realize that things are thought differently in Italian and acknowledge it. 4. You will find that although in English, you would always need a pronoun, it's optional and almost never used in spoken Italian. The reason for that is that unlike in English, the verb in Italian changes form according to the person it refers to. For example, the verb in the infinite form for 'ear' is mangiare, this takes the form 'mangio' if you are talking about yourself, 'mangi' if you are talking to second person and 'mangia' if you are talking about a third person. so just saying the word 'mangio' is equivalent to saying 'io mangio' - I eat. 5. Gender- Unlike English, Italian (and many other languages) assigns gender to inanimate objects as well. This might sound a bit odd to native English speakers, but this is important in Italian and may be a bit difficult to comprehend. Although, you may not worry about the gender if all you want is a brief knowledge of Italian (people will understand if you do a mistake with gender but still may be able to understand more or less what you mean) . For example, a ticket has a masculine gender - 'biglietto'. Generally speaking, words that have 'o' in the ending are masculine and ending in 'a' are feminine. Correspondingly, the plurals of these words end in 'i' and 'e'. Just be careful of one more thing, some words that have 'e' in the ending may also be singular masculine or singular feminine. So, be cautious while applying the reverse logic of 'e' to plural feminine. 6. Articles- Although, the articles are not so often used (or skipped sometimes) in English, they are very important in Italian. What does complicate the case in Italian is that the article changes according to the gender and the number. The basic articles being - un (a), il (the) for masculine and una (a) and la (the) for feminine words. Since, this is not supposed to be an extensive review of Italian, I will (and have skipped) skip a lot of complications. There are lot of variations in the articles as well. 7. The distinction between declarative sentences and interrogative sentences is not very clear in Italian as it is in English (also because the pronoun is usually skipped). What I mean by that is that in English the sentences : I can have an apple (declarative) and Can I have an apple? (interrogative) are clearly distinguishable because of the order of pronoun. However, in Italian the same sentences would be: Posso avere una mela (I can have an apple). Posso avere una mela? (Can I have an apple?). Basic words/phrases (of course every learn Italian video or book or course has it). What I will provide is explanation of when to use them: 1. Ciao - Hello and Bye. Use with care. This is an informal way of saying Hi. So, probably won't look nice if talking to an older officer or a complete stranger a lot elder to you (or alternatively if you use a very formal way to talk to a young kid, it might sound very funny to the Italians). 2. Buon giorno - Good day. This is a very useful word. This is sort of the safe word that I have been using for some time. It can be used in place of Ciao to almost anyone - young and adult. This is also the word people usually use when one enters a shop. It is polite to great people and you will find it prevalent everywhere. 3. Buona sera - Good Evening. The substitute for Buon giorno. in the evening. Same usage as buon giorno. 4. Posso avere un biglietto?: Can I have a ticket? I am not sure if this applies to other places in Italy, but in Genova, the tickets for local buses (and train within city limits) are purchased from newspaper/tobacco shops. If you would like to travel somewhere via a bus or other means you can use this statement for asking for a ticket. 5. Grazie/Grazie mille: Thanks./ thanks a ton!! A possible scenario: Buying a ticket You: Buon giorno/ Buona sera Vendor: Buon giorno/ Buona sera (Alternatively: Dimmi - tell me) You - Posso avere un biglietto di A.M.T. , ordinario? (Alternatively : Posso comprare(buy) un biglietto? in genova AMT is the main agency for tickets and ordinario is the ticket type). Vendor - Si (gives the ticket) Vendor - uno e sessanta (1.60 euros) You - (giving money) Grazie Vendor : Grazie, Ciao. (Alt Arrivederci - Good bye!) You : Ciao/ Arrivederci In a supermarket while billing: Supermarket employee: hai tessera? (do you have a card? - like a loyalty card) You: No. Supermarket employee:: sacchetto? (a plastic carry bag?) You : No. Misc. : Dov'è il bagno? Where is the bathroom/ toilet? Permesso: Asking for permission to go through. This is used when there is someone blocking the way and you would like to go through. scusa/scusi : Sorry (formal/informal). use Scusi in formal situations. these words are used in cases where you are in a hurry and you accidentally hit someone. Vorrei: I would like to. For example : vorrei una pizza (I would like a pizza) . non parlo Italiano: I don't speak Italian. okay: ok va bene: Alright. Quant'e' : How much? ( after purchasing a meal at a restaurant ) Some of the resources that I find useful: 1. Duolingo : a very nice app for learning different languages. 2. italian-verbs.com: a list of a lot of Italian verbs and different forms according to person and tense. Some of the useful videos : 1. This is one of the most useful resource I found, it gives a good ideal of Italian: http://www.youtube.com/playlist?list=PLKICLEvD2DfpUtjb4nc0Vr9QfdYn_Ursj I will keep adding more in the due course!! Recently, I had to place an order for a work desktop (and of course you all know that I work with neuroimaging data). So, I had to specify the minimum requirements that I needed in the system. Anyone who has done something like this before would understand the complications (and the joy) associated with this.
There are a number of parameters to choose from especially when you don’t have a very strict restriction of money. This is complicated by the fact that there are no guidelines that tell you how to proceed. Since, I went through this (and had nothing else to do), I decided to write this post about how I went ahead with the specifications. Note: Following are mere guidelines!! First of all, you need to ask certain questions. 1. Type of work: What kind of data processing do you require? Does your work include basic fMRI processing (resting state and task related), structural analysis, DTI analysis or connectivity analysis etc. Certain kind of analysis would require more processing and RAM while others don’t. My work involves basic task related analysis of fMRI data as well as some structural analysis. 2. Amount of Money: How much money are you allowed to spend? This might sound trivial, however, money is a very big constraint and you would preferably try to reduce the costs (I will give both high end as well as less expensive options both of which should work fine). 3. Assembled or not? Do you have the required technical skills to assemble a desktop from basic parts? In most cases, this may not be valid, as most of you would have a purchase/technical store that would get assembled computers from manufactures. Getting an assembled configuration gives you lesser choices but is already optimized by the provider company so that you don’t have to worry about compatibility issues. So here are my recommendations. I believe it’s always better to get a system better than your current requirements (as they are sort of investments and are not bought frequently). The three things that you need to pay the most attention to are : CPU, RAM, Graphics card. CPU: There are two major providers of CPU- Intel and AMD. I personally don’t have experience with AMD processors. Intel produces a large variety of CPU processors and there are a number of them to choose. The most famous being the i series processors for laptops and desktops. What you should prefer, however, are the Intel Xeon processors. These are workstation processors designed for stability and continuous work (which I believe most of the neuroimaging folks would relate to!!). Xeon processors are classified according to the series (no surprises there!!). E5 I believe are the latest (as of March 2014). CPU benchmarks (passmark) is a great tool to know how the processor you have chosen performs compared to other processors in the market. You should also check their price performance (compared to price how good they are). My choices for the CPU: 1. Intel Xeon E5-1660 v2 @ 3.70GHz 2. Intel Xeon E5-1650 v2 @ 3.50GHz 3. Intel Xeon E5-1650 @ 3.20GHz 4. Intel Core i7-4930K @ 3.40GHz RAM: RAM is a critical component of the system and you would require lots of it especially for neuroimaging analysis!! Generally, the more the better. A RAM of around 16 GB (preferably divided as 8x2 or better 4x4 ) should work for most of the analysis. Also, it’s important to get the latest and the best in the technology. I would suggest getting DDR3 (third level of RAM technology) at 1866MHz. My choice: 1. 16 GB DDR3 1866MHz Graphics Cards: Many researchers would say that graphics cards are not necessary component of a neuroimaging system. However, recent advances in image processing and use of heavy visualization softwares requires using atleast a mid level graphics card. It’s important to note that while many of us would be tempted to get the best graphics card from GeForce or Tegra or Tesla (mostly so that the system can satiate our gaming needs as well!!) , Quadro are the preferred graphics cards for neuroimaging analysis. These are designed to be workstation graphics cards and are stable and can work continuously for hours. My choices: 1. NVIDIA® Quadro® K2000 2GB GDDR5 CUDA core 2. NVIDIA® Quadro® 2000D 3. NVIDIA® Quadro® 600, 1 GB, 96 CUDA cores, There are other requirements that are not so critical like hard disk, operating system, monitor. But they depend more on your preference. I would recommend 2 mechanical Hard disk 1TB or 2TB each (SSDs are new, not so stable, have had issues with Linux!!). I personally prefer Ubuntu 12.04 (Long term support) and a high resolution 23 inch (1920x1080) monitor. I took considerable help from these additional resources: 1. http://www.reddit.com/r/neuro/comments/1f94nu/building_a_pc_for_neuroimaging_data_analysis/ 2. https://sites.google.com/site/mritutorial/iii-neuroimaging-workstation-tutorials/tutorial-1-hardware-recommendations (has suggestions for dell based systems and other suggestions) 3. http://www.youtube.com/watch?v=LV3YBhrD_D4 4. http://www.newegg.com/ 5. http://cpuboss.com/ 6. www.nvidia.com/content/quadro_fx_product.../TB-04631-001_v04.pdf Writing a research paper is difficult, very difficult. I found it out the hard way when I had to write my first research paper. Reading a paper is very easy but writing something that you have done (without copying and pasting from similar papers (BTW also called plagiarism)) is very difficult and you have to be very innovative but at the same time sound technical, clear and interesting. After I actually wrote the paper (which took me and my colleague around 6 months), the way I look at the papers completely changed. Now, when I see certain portions of the published peer reviewed papers, I know that this section was completely due to the pressure of the reviewers, this section was because the authors wanted to highlight certain aspects (and hide some others). So, now you would be thinking, why have I started this rant of mine? Good question! It is because there are certain things that have really annoyed me about the papers which might be inherent problems of the field. But still annoying. Specifically, I am upset about some of the practices people do and the way they write their methods section. 1. Talaraich Co-ordinates. I do respect the idea and effort that was put in making the Talaraich atlas. But, now I believe if you really want to confuse people and don’t want your study to be replicated, you use these co-ordinates. I know that there would be people who would disagree with me, but I guess this is my opinion. First, there are well known issues with the Talaraich atlas (from a single subject etc.). A lot has been written about it, so I won’t elaborate here. Second of all, there is no single accepted way of transforming it into MNI. There are ways - Matthew Brett’s script and lot of others (see http://www.sdmproject.com/utilities/?show=Coordinates). But mostly they are empirical and approximations. So, now I just get annoyed when I see Talaraich coordinates. 2. Cluster level correction: The field is well aware of the issues of multiple comparison and a lot has been said – some in a funny way and some in more serious ways. However, the issues has now taken a different direction and incorrect or less understood ways of correction have come up. One of them being the Alphasim/ monte carlo simulation method which is often poorly described/incorrectly used in the studies (see Bennett, Wolford, & Miller, 2009). I am not saying that the method is wrong. But the way it is written is often unclear (probably even to the authors). This leads to misleading statements like ‘The results are corrected for multiple comparisons’ in the papers when they are corrected at cluster level and not on the voxel level. Also, FDR voxel level has been proven not to be appropriate for inference on peaks or regions of activation than the cluster level FDR (Chumbley & Friston, 2009). SPM software has already stopped showing results at FDR voxel level. However, FDR voxel level are still prevalent in the papers. 3. Results on Brain slices. I know that a lot of people will disagree with me and say that the way of presenting your fmri results is upto you. But I would argue that if you really like your results on brain slices, put them in supplementary or next to a whole brain image. What brain slices do, is that they let u see only what the authors want you to see which I believe is against the idea of a results section. The results section should have all the results as it is without the interference from the authors – their way of thinking or their beliefs. I know that this might be difficult to take out from the paper, but we can try with results that just don’t show one slice of brain! So now, whenever I see fmri results on a brain slice, I start believing that there were probably some activations, that the authors didn’t want me to see and more often than not, when I see the results table, there are other regions which are activated. Lot of times hiding other activations might not be the aim of the authors but to showcase specifically the activations, however, often what this leads to is a bias in the mind of readers (quite frankly how many of us go through the tables for list of areas activated in each task unless we are looking for co-ordinates or specific regions). The reason I stress on this is because, what this eventually leads to is strong ideas in the field about the task and if the paper is influential enough, to studies that tend to approve the regions in the studies forming a vicious circle questioning studies that show other regions activated in similar tasks. 4. Short methods sections. This is very true of some journals that simply think that methods are not so important, it’s the results that are critical. I believe that methods are almost as important as the methods. For example, there are multiple ways to make a something that looks like a cake, but all of them might not be edible. 5. Previously used methods: The sure shot way to ensure that your results are never replicated ever – “The protocol previously used in the papers (1 2 also see 3 ) was used for the data”. Most often the papers 1,2 as well as 3 refer to different protocols which are somewhat related to the current paper but possibly use completely different parameters (which is not very obvious). A reader only interested in the fascinating results might not really bother about them but people who would like to replicate the experiment would be sent on a treasure hunt to find what parameters were actually used. Strangely, the journals do not mind such statements. 6. Methods in books. This is another way to ensure experiments are not replicated. “The values were calculated using the criteria X (reference to a book)” Often this book is a textbook, not freely available and even if u manage to get your hands on it, you are on another journey to hunt the location of the criteria in the book. The reporting of methods is improving however, with efforts like pre-registration of studies and other practices of reporting results. Hopefully, this will improve further in future. Consider the case where you are gone on a trip to a foreign land where u know no one. One fine morning you go to the local market and in the rush of the market someone steals your wallet/purse. You have no money, no passport and no one to contact. You don't know who to approach for help. You don't know the language and you want to contact someone for directions who would you approach to ask for help and who you would like to avoid. Would you approach someone who looks untrustworthy or would you avoid that person??
I recently gave a presentation on a topic related to this. It was based on the 2011 cerebral cortex paper "Endogenous testosterone modulates prefrontal-amygdala connectivity during social emotional behavior.". Its an open access paper and you should give it a read. The paper contributes to the prevailing ideas of influence of neuro peptides on social behavior (the prominent being the effect of oxytocin on trust). The current paper discusses role of endogenous testosterone in approach/avoidance behavior and how it modulates the connectivity between amygdala and prefrontal cortex. Steroid hormone testosterone has been shown to play important role in the regulation of social emotional behavior. Studies have implicated testosterone in reduced human trust as well as in reduced lying, aggressive behavior. Past studies suggest that we can make trustworthiness judgments from faces in less than 100ms and that peptides like oxytocin and testosterone modulate our judgments. Trustworthiness judgments are suggested to be a prerequisite to approach/avoidance behavior. While the paper does present interesting results that might contribute to understanding human social behavior, the paper does lack some of the rigorous statistical standards. For example, all brain maps (brain images) are not at corrected thresholds. Much has already been written about the necessity of corrected threshold (see Craig Bennett's salmon paper). Although the results say a corrected threshold of FWE p<0.05, the analysis is not whole brain but for selective regions. The instructions provided to the participants was also not very clear. The paper relies on the idea that happy faces promote approach automatically. The participants were not told to specifically approach or avoid a face but an irrelevant top down task of approaching or avoiding happy faces (and vice versa for angry faces). This might explain the activation of VLPFC/FP in the results. Also, the use of salivary testosterone may not be a great predictor of testosterone in the brain. The specific use of male participants has been accepted as a limitation. Overall a paper with limitations and less rigorous statistical measures. Find my presentation here. It may not be a great achievement, not even an achievement at all. But it is a sort of moral booster. I guess there have to be events of personal joy where you take a rest from the depression of the research work and enjoy a bit of success (how small it may be). The extension is a work of my getting fed up of the laborious work of using aal (Automated Anatomical Labeling atlas) every time I had to find out what regions are activated in my activation map (although I have become better than before in distinguishing Prefrontal from Parietal!! now). The extension although was written in 2012 but I was too lazy to make documentation or make it organized (nor that it is organized right now). However, it's better now and I feel that atleast someone might find it useful to automatically batch aal results and print a copy for later retrieval.
Well, I almost forgot to add the link : https://github.com/ateshkoul/NeuroImaging/tree/master/spm/batch_aal or you can download it as a zip here: https://github.com/ateshkoul/NeuroImaging/blob/master/spm/batch_aal.zip The script is on github. I believe all our programs and scripts should be there for everyone to use and modify. The extension is easy to use and install: 1. If you have the zip, extract the contents to a folder batch_aal 2. Copy it to your '....spm/toolbox/' folder. (if on windows something like 'c:\Programs\spm\toolbox\batch_aal') 3. Use it from SPM -> Tools -> Batch AAL. Thats it!!! |
About meI am a neuroscientist working on social cognition. (I was told not to be fancy.) Archives
June 2016
Categories |