Here’s an official announcement that I will no longer be maintaining this blog. ‘Official’ because the blog hasn’t seen an update in a long time. This is just the full stop that ends the sentence.

Blogging has been a valuable exercise for me. An exercise that started during my postdoc in St Louis, and continued through setting up and  establishing my own lab in St Andrews. I always intended it to be a series of notes to myself that I could make public for others to benefit from. This, from the first post in 2010:

I hope that this blog helps me to gather to one place many of the bits and pieces, both on- and off-line, that have helped me over the past couple of years.  I also hope that I get into the habit of updating it in a way that is useful when I need to find out how I tackled a particular problem in the past, or even when I simply need to find a web-site I know was previously useful to me.  If it proves to be useful to anyone else in the process, then that will be a marvelous outcome too.

There are still times when I google an SPM error, to find the top hit from my own blog on how I overcame it years before. Judging by the emails I also still receive about historic blog posts, I think the blog has served its purpose of being useful to others. Marvelous.

You’ll still find me on twitter: @akiraoc, and I’ll keep updating the various non-blog pages on this site. Here though, I’ll end with some posts that, for all sorts of different reasons, remain particularly important to me.

Thoughts on Platt’s Strong Inference article from 1964

A reflective post on how I learned to give better talks

Guidance for students on how to email your lecturers

Documenting how to stop participant attrition in online studies

Why the Martingale System for winning at roulette (or any game) doesn’t work

A guest post by Radka Jersakova on conducting online studies


Last weekend I had the honour of being Best Man at a university friend’s wedding. It was a beautiful day spent in the sunshine of St Albans and then in the low-ceilinged, close comfort of the oldest pub in England, Ye Olde Fighting Cocks.

Ye Olde Fighting Cocks, St Albans
Ye Olde Fighting Cocks, St Albans. from

As Best Man I had a certain number of duties to carry out, with the speech amongst the most highly anticipated by those attending the celebrations. For those unfamiliar with what is expected here, the Best Man’s speech is traditionally the last of the speeches and the point in proceedings when thanks and sentimentality give way to humour and a raucousness that sets the tone for the night ahead. For weeks beforehand, people had been asking how I was getting on with it, noticing my terse response (“getting on just fine thanks”), and reassuring me that there was plenty of material to draw on. The expectation that it be funny was pretty inescapable.

I lecture statistics to 150 students on a regular basis. Over the past few years I have managed to overcome my hatred of public speaking and relax into these one-sided conversations on t-tests, ANOVAs and regression. One of the luxuries of speaking in front of large audiences as part of my job is that I know what it feels like and I know how to deal with the mechanics of getting my words out of my mouth in quite an intimidating situation. There are even moments in lectures now when I notice that I’m in a flow state, enjoying the fluidity of speaking about something I know well. For this reason, the prospect of getting up in from of over 100 boozy partiers and speaking about a good friend was not what I found intimidating. The expectation that I make them laugh, now that was scary.

I know that to speak well, I need to prepare (I have written about the routine I go through for important talks here). This is exactly what I did for the Best Man’s speech. The result was one of the most exhilarating experiences of performing in front of an audience I have ever had – the audience enjoyed themselves and I had a tremendous time. I didn’t have to buy a drink for the rest of the evening! Here is what I did to get into that position.


Be Yourself

1. I reassured myself with the knowledge that, in standing up in front of lecture theatres full of students,  I am paid to do something very similar to this. The major difference between lecturing and speech-giving was what the audience expects of the content. I knew that I was expected to offer toasts to thank various people, but whom exactly? To help with this googled the running order for content within a Best Man’s speech. By chance I found The Art of Manliness’ 10 Steps to the Best Best Man Speech, from which I got some suggested running order information, but much more importantly, I was reassured by the insistence that I ought to be myself. I have never wanted to stand up in front of people to make them laugh, but I do nonetheless enjoy making small groups of friends laugh when telling stories in the pub. Bearing this in mind allowed me to feel comfortable in not trying to ape my favourite comics, but simply allowing myself to find my inner story-teller and let him speak to a larger group of friends. This was the me I tried to be when writing the speech in advance of the wedding.


Write and Practise the Speech in Advance

1. Write the speech beforehand. Write it out even if you’re not going to read it. I always write out important talks so that I can practise my phrasing and I edit them to whatever worked best after each run-through. Thus, I start with a script in rather broken spoken prose, which is edited into something that sounds natural by the time I’m done with it. Over the course of practising I learn what I want to convey in each sentence so that I can say it  in any number of ways, off-script, by the time I get to delivering it to an audience. I know that this is a matter of personal preference, but this is what works for me and I could never deliver any speech or talk without practising it a few times first.

2. When lecturing or giving talks, the transitions between slides are often tricky points. My Best Man’s speech had similarly tricky transition points where I moved from toasts to anecdotes or from one story to the next. Scripting these transitions as part of scripting the entire speech gave me an idea of how to move on as seamlessly as possible.


Work on Timing

1. Don’t out-stay your welcome. I can’t over-run my lectures because students will start leaving. They have other places to be. Weddings guests probably won’t leave, but they won’t applaud you for going on and on either. Silky, a stand-up comedian attending the wedding spoke to me as we were sitting down to dinner. He gave me the following advice: “If it’s going badly, get off quick. If it’s going well, get off quick.” In other words, keep it short. Before I started writing the speech I was aiming for about 10 minutes. Run-throughs lasted about 13 minutes (an acceptable timeframe according to the groom, whom I had asked about this beforehand). If you’re running to a tight schedule, practising the speech will give you an idea of whether or not you need to remove content.

2. Comic timing is a little harder to work on. This is something I have rarely had to worry about in lectures (I tend to play them straight) and I’m not sure how I would go about practising comic timing other than by doing this sort of speaking more. Something that threw me off a few times was that people started tittering before I had delivered the punchlines. The audience expect you to be funny and they want you to feel comfortable, so they will laugh when you give them an excuse to. This made me fluff a line or two. It is something I will be more mindful of should I ever have to do this again.


Logistics and Planning

1. Know your AV equipment. I delivered the speech into a hand-held microphone. I had seen the first speaker struggle a little with microphone distance so I was determined to be careful of making the same mistake and, in the end, delivered my speech with the mic resting on my chin just below my lower lip. It probably looked weird but I managed to get through the whole speech without any microphone dropout. (In future I will have a go on the amplification equipment beforehand so I can work out something a little more elegant.)

2. Coordinate toasts and readings with other speakers. When you are delivering a lecture course, you want to avoid not covering important material (dangerous for exams) and duplication (boring). The same is true of wedding speeches. The night before, the groom and I had discussed who was giving which toasts so that, by the time I had finished, everyone who needed to be thanked would had been thanked. Had we not had this conversation, the groomsmen would have gone without a toast – an omission I’m glad we avoided. On a related note, I also opted to read a short passage to the bride and groom to close my speech. It was only at the wedding service when one of the passages I had been considering for my own speech was read that I realised I had got lucky the choice I eventually made. If you are doing something unconventional like reading a passage during your speech, have a quiet word in the groom’s ear to ask what readings they have planned for the service well in advance.


Being Funny

The three points below target a specific aim, being funny, which isn’t a priority for me when I lecture. I don’t make much reference to lecturing below because these points are specific to my experience of the wedding speech situation.

1. Despite the pressure, being funny is not the be-all and end-all. Having typed “Best Man’s Speech” into Google, I was surprised to find  the first auto-complete suggestion to be “Best Man’s Speech one-liners”. I don’t use the slides my course textbook publishers give me when I lecture, and I would be similarly wary about using jokes about other people to portray my relationship with the groom. What I wanted to do above all else, was to paint a picture of the groom as I know him. Having said this, such is the pressure to be funny that I’m not surprised people google jokes for use in wedding speeches, or end up telling embarrassing stories from the stag do. Again, the  Art of Manliness article delivers reassurance:

What gets people in trouble is attempting to be funny by sharing some embarrassing story or cracking some lame joke about a ball and chain. It usually comes out horribly and no one laughs. It’s okay to share a humorous anecdote, but not one that gets laughs at the expense of your friend and his new wife and embarrasses them and their guests.

This advice set the tone for the stories I wanted to tell. I wanted those in the audience who knew the groom to see him in the jokes I was telling and for this recognition to be funny in and of itself. I also wanted to capture a range of experiences I had with the groom, from those that were funny to those that were sad. The sadder moments would act as points from which to rebound back to laughter, but would also help the audience understand what a lovely thing it was for the groom to have met the bride at the time he did.

2. Avoid in-jokes. My university friends would invariably ask if I was including their own favourite university story of the groom. Many of these stories were very funny, but only in the context of the many in-jokes we shared as a group of close-knit friends. I largely avoided these  references because I wanted to appeal to as many in the audience as possible. Those that I did include were equally viable as terrible puns or cultural references, which the university crowd found funnier because of their shared history of appreciating them.

3. Enjoy the format. I was in the privileged position of speaking to an audience who expected me to make them laugh. When writing the speech I experimented with jokes, tweaking wording, timing and structure. I eventually settled on a narrative that in the end called back to humorous stories about the groom to illustrate how normal the bride is in comparison. I have seen comedians using and even explaining this device to great comedic effect, and incorporating it into structure of my own speech gave me a sense that I had actually written a funny speech. This is undoubtedly an aspect of the Best Man’s speech that I would not have thought to focus on had I been preoccupied by the prospect of public speaking. My experience of lecturing allowed me to build on its commonalities with giving a Best Man’s speech to embrace and ultimately enjoy the format tremendously.

Frontiers use a pre-publication peer review system that differs from the standard peer-review system used by most traditional journals. After undergoing Independent Review, an initial peer review in which reviewers give star ratings and make comments on your manuscript (the standard reviewing format), your manuscript moves to Interactive Review, an online forum in which points from the Independent Review are dealt with by the authors and there is back-and-forth discussion all parties are satisfied. Frontiers make much of this system, claiming that it increases reviewing efficiency and leads to an average submission-publication lag of only 3 months.

A paper on which I am an author is currently undergoing Interactive Review. Up until now I can’t say that the review process has been particularly speedy, with some reviewers reluctant to engage in the Interactive Review (indeed, I suspect many of the time-savings from submission to publication come after the manuscript has been accepted for publication). Frustrated, I emailed the Frontiers in Psychology Editorial Office to ask what their official guidance is on the time it should take authors and reviewers to complete each stage of the  review process, and how they deal with tardiness. Here are their answers:

  • Independent Review – 10 days
  • Interactive Review, authors’ responses to Independent Reviews – 45 days
  • Interactive Review, reviewers responses to authors’ responses – 10 days
  • Interactive Review, authors’ responses to reviewers’ responses – 10 days (and so on)

Frontiers say that they “have a dedicated team…  who work on ensuring that the review process of manuscripts runs smoothly. Should participants become delayed, we monitor the situation and also remind them about taking action.”

This sounds like a good system in principle, though I remain to be convinced about how effectively delays are dealt with. It certainly seems that if a reviewer wants to slow down publication of a paper, they can do so at little cost to themselves. Just like traditional peer review, Frontiers’ review system relies on goodwill from all participants and a strong editor, maybe even moreso as there are any number of points at which a reviewer can bring the Interactive Review to a halt. Apart from the increased transparency (it’s blindingly obvious that Reviewer 2 not only hates your paper, but can’t be bothered to say this quickly), there doesn’t seem to be much that is revolutionary here. I suppose I’ll just have to wait and see.

I spent most of this summer in St Andrews writing research papers. This prolonged period of writing gave me time to consider the publication of my own work along lines I haven’t fully considered before. I was able to think not only of the quality of science typically reported in the journal to which I was considering submission, but also of that journal’s publication model. For the first time in my career I felt it not only desirable, but also sensible, to submit to open access journals. It’s not that I haven’t wanted to publish in open access journals before, it’s just that there have been too many things stopping me from breaking with the traditional journals.

Open access image via salfordpgrs on flickr:
Open access image via salfordpgrs on flickr:

So what changed? Of course, the traditional publishing houses have had a lot of bad press. Their support of the ultimately unsuccessful Research Works Act, set against soaring profits and the unsavoury business practices of academic journal bundling all demonstrate how committed traditional publishing houses are to making money, not increasing access to research. More personally, PubMed’s RSS feeds for custom search terms (informing me as soon as something related to ‘recognition memory’ is published), Twitter’s water-cooler paper retweets and Google Scholar‘s pdf indexing, mean that I usually learn about and can get access to articles I am interested in, without needing to know where they are published. Over the past months and years, subscription-model journals have started to feel old-fashioned, maybe even willfully so. It’s now the case that if university scientists are interested in my research, they will find it regardless of where it is published. If the public are interested in my research, whether or not they will be able to read it depends entirely on how it is published.

That said, Google Scholar would be useless as a source of papers if researchers and universities didn’t make pdfs available for it to index. It’s here that the work university libraries do to promote open access is crucial. At St Andrews, we use the PURE system which makes green open access – uploading author versions or publisher versions usually after an embargo period –  straightforward. Beyond this though, the Open Access Team frequently encourage us to provide this sort of green open access. For example, earlier this week one of the open access librarians tweeted me to tell me that I was entitled to upload the final version of a paper I had been holding off on uploading. In doing this, they cultivate an environment in which providing open access is seen as a responsibility we have to those who might want to read our research.

While green open access can work, it requires efficient management. The St Andrews Open Access Team seem to have a million publisher-specific checks they have to make before they will allow a pdf to go into the Research@StAndrews:FullText repository. Surely gold open access – publishing in journals whose business model doesn’t involve protecting access to their outputs – would make things much easier. The one problem with gold access, even from a the point of view of a researcher who wants more than anything to publish in this way, is that it is expensive, really expensive. A paper in PLOS ONE costs $1,350, Frontiers, €1,600 and Springer Plus, £725 (though this all may change with PeerJ’s author subscription model). Of course it makes sense. A journal that doesn’t charge subscription fees needs to recoup its costs by charging to publish. And here’s where we run into major barriers to the uptake of gold open access. First, gold open access publishers are asking universities to spend money to publish their own researchers’ work when they’re already spending an eye-watering amount on accessing work that the same researchers have previously published. Second but maybe more importantly in terms of journal submission choices, gold open access publishers are now forcing researchers, not their university libraries, to face up to the costs of publication.

That I, not the head of my library, must think about how to fund the journals that publish my research goes against the traditional subscription model of academic publishing. Moreover this financial division, and the problem it poses to open access journals, almost certainly exists at every single university in the UK. In an ideal world, I should be able to dip into the library’s subscription budget every time I publish in a gold open access journal. If all researchers knew that their submission to Frontiers in Psychology wasn’t jeopardising their travel to next year’s conference in San Diego, gold open access would be set. It’s only when universities recognise this, that gold open access publishing payments should come from the same pot as journal subscription payments, that open access publishing will take off.

And so to why I was able to consider submitting to a gold open access journal. The St Andrews Library Open Access Team have a fund specifically for gold open access publishing. A cheeky twitter request as to whether they would support my submission to an open access journal was all it took for me to get the thumbs up. Together with the green open access resource at Research@StAndrews:FullText and maintenance of the existing closed access journal subscriptions (for now), the gold open access fund helps to provide the full range of publication options for St Andrews researchers. It’s a comprehensive approach to open access that makes me proud to work here.

The process of rotating data in Excel, such that rows become columns and columns become rows, is pretty straightforward. Copy, and then right-click on the destination and select the ‘transpose cells’ paste option.

Right-clicking to transpose data.

Things get a little more complicated if you want to transpose a series of cell references or formulae e.g. “=A14” or “=NORMSINV(A14)-NORMSINV(1-B14)”. If you don’t have all your cell references in absolute format, Excel will get the transposition all wrong. One way of getting round this is to find and replace (CTRL-H) all your = signs in the array you want to transpose, with a symbol that Excel finds meaningless, like #. You can then copy and paste-transpose your # cell references, and once you find and replace the #s with =s (in both your original and transposed arrays), you’ll have achieved the transposition you’re after.

In A, the transposed cell reference gets messed up (Excel transposes the direction of the reference during the transposition). In B, CTRL-H has been used to find and replace all =s with #s. The cell references look to have transposed correctly. The correct transposition is confirmed in C once CTRL-H has been used to re-replace all the #s with =s.
In A, the transposed cell reference gets messed up (Excel transposes the direction of the reference during the transposition).

In B, CTRL-H has been used to find and replace all =s with #s. The cell references look to have transposed correctly.

The correct transposition is confirmed in C once CTRL-H has been used to re-replace all the #s with =s.



Tomorrow morning I fly to SfN 2012 in New Orleans. There has been some turbulence in the immediate lead-up to it – an unscheduled flight to Ireland putting back my departure by a couple of days and a trip to A&E with anaphylaxis being the major ones- but nonetheless, the trip is happening.

This is a pretty big deal for me and for the lab. It’s the first major presentation of data we have collected independently of more senior PIs (and they’re not bad-looking data). It’s also only the second time I will have presented at this particular conference, which can be a tad overwhelming. This could be the start of something good.


Of course, there’s also some more social fun planned. I’m looking forward to the #SfN12 #tweetup (Monday night), when I’ll be putting some real-life personalities to some online ones, and the Scottish Neuroscience Group drinks (Tuesday). I’m not anticipating any more storms, other than the odd hurricane, rum and all.

Image via Wikipedia

Here are a number of free online sources of music and background noise that are particularly good for high-concentration tasks.

And here are a number of (not free) albums I find particularly good for this purpose.

This isn’t a comprehensive list by any means (that sort of thing can be found on Lifehacker, as you might expect), but over the past few years, I have tweeted and blogged about this a bit, so I’m simply tying up some of the odds and ends in one place.