Tags
affectogram, analysis, bipolar, data, mania, mood charts, mood swings, Moodscope, triggers
In addition to the Triggergrams I recently discussed, Moodscope Plus has another data display: Affectograms! “In one simple chart which takes the form of a ‘heat map’ you’ll be able to see at a glance how your scores for all twenty emotions have varied over the past 28 days.”
This is the report that I wanted from Day One of using Moodscope. There’s a lot more data being recorded than just your daily score, and having access to the details has been very enlightening to me.
My Moodscope Affectogram for late February 2012 through late March 2012
I think this is a pretty intuitive graph, and the Moodscope team have done a brilliant job of including all the information you’d need to put the data in context. It even shows the numeric score and your notes when you mouse-hover over the scores in the chart.
So here is my reading of this graph:
- Strattera was really bad for me, and led to low positive moods in late February and early March. My worst days on March 5 and 6 suddenly reversed when I quit taking Strattera on March 7. March 6 would have been a dramatically lower score had I taken the test at noon, possibly my lowest ever, due to a severe medication-induced panic attack that day.
- March 17 was a gray day and the first time I’d gotten 8 hours of sleep in 2 weeks. Things improved after returning to a little less sleep – 7.5 hours seems to be my ideal.
- I suffer more from lack of positive affect rather than presence of negative affect. It’s not that I feel bad so much as I just don’t feel good, though I may be taking negative emotions as the status quo. This might have been quite different prior to starting mood stabilizers, which reduce both the bad and the good. Comparison over time would help me learn how well I’m evaluating negative affect.
- I had a short manic episode from March 21-24 followed by some irritable hypomania. That was partially triggered by turning in my dissertation manuscript, and I also noticed that it was not only unusually sunny and warm that week, but I was outside to enjoy it. The change of seasons is a known factor in bipolar mood swings for many individuals, and this suggests it’s a much more substantial trigger for me than I realized.
- Irritability should be rated higher much more frequently, but I often can’t tell until later in the day (or even days later) that I was irritable.
- “Attentive” ratings have little to do with how distracted and/or productive I actually was, probably because these scores are taken before my ADD meds! The dates rated lowest were also when I cut my Concerta dose to make the supply last longer. The days rated highest were the most manic dates, when I was actually most distractable, did not take ADD meds at all, and obviously had the least insight. I was likely conflating alertness with attentiveness. I generally have a very hard time rating my attentiveness, as my early Moodscope scores would show: I consistently rated it low because I’m simply inattentive by nature–and hadn’t taken my meds yet for the day.
- I’ve been really inactive lately due to 12 hour days at the computer finishing up my dissertation. This is also something I should rate at the end of the day. I was much more active on March 21-25 than the score suggests, and you can see a loose correlation between mood and physical activity when I add in my step count graph from Fitbit: for example, March 8. Causality is not implied, but I suspect it goes both ways.
My step counts for late February through late March 2012
These data suggest that I can get a better representation of the day if I take the test at the end of the day. This was even easier to see since I often add my notes for the days post hoc, and they tend to be more consistent in terms of overall objectivity. So I’ll be experimenting with changing my mood tracking habits to see if an evening score is more useful than a morning score.
One detail that would help me see trends more easily would be reversing the order of the positive emotions. Currently the lows are shown on top, but it would be more logical for the highs to be shown on top. Mirroring the negative emotions (as they are currently displayed) is also more sensible. This would group the negatives (darker shades) together in the center and the positives (lighter shades) at the outside, so the bars and percentages at the right would become more useful too. This would be much easier to read and interpret appropriately. The daily strength on positives should therefore move to the top to maintain the visual flow. Although the summary would then precede the details, it shouldn’t be a problem for reading the graph and would be a worthwhile tradeoff.
The main new feature I’d like to see (other than data download, image export, and printing) is adjustable date ranges. A larger range of dates would let me compare time periods and do much more detailed analysis. I want to see what this would have looked like last year, for different treatments, and during known mood episodes, as the picture might look quite different.
Don’t get me wrong – I simply love the Affectograms! They are very helpful for better understanding what is really influencing my moods more specifically, and I think they would even be useful to take to my therapy appointments. The Affectogram shows me when situational trends are distinct from specific triggering events, so I have a better sense of the different influences on my moods. That helps me better prepare for and handle these situations in the future, or at least accept that the challenges I may be facing are in fact situational and not my fault. Kudos to Moodscope for this awesome feature!
I gotta check out this Moodscope…
But I am terrible at keeping up with these things.
The easiest way is to make it part of an existing habit, e.g., do it when you take your meds or brush your teeth or have your morning caffeine. Some people set alarms on their phones but I think making it part of daily rituals is better. There’s an optional daily email that serves as an inbox-to-do-prompt for me, which is good because I’m trying to re-jigger the timing for doing this and breaking/re-establishing ingrained habits is really really hard for me.
The score-sharing might also be helpful for the Clown when it comes to managing things as a daring duo. It helps Mr. Chickadee know when to keep an eye on me for crazy behavior, which is good because I don’t always recognize it even with the warning of a wacky score.
I’d never heard of Moodscope before. Thank you for sharing about this. I will certainly be paying them a visit soon.
I’m a fan – this is my 4th post about Moodscope! But the last one for awhile, most likely. I think it’s a very handy and easy to use tool, though there are lots of options for mood charting.
I don’t get it. How do I read the affectogram properly? If I look at this moodscope affectogram visible here and I try to see what’s been going on for you is it correct to read that you were feeling bad about the ‘active’ measure for the first 6 day’s of the graph? You’ll have to explain it because I just don’t get it.
Red = bad and yellow = good on both scales (the color key is at the bottom). The overall score is cumulative: the first few days of this diagram had a lot of red and dark orange for the positive moods, which is basically bad, especially when taken together. You can see from the line graph at the top that I had overall scores in the 30′s and 40′s, and that’s pretty lousy for me.
But the other thing I like is that this graph shows me that sometimes one factor is going really well or poorly when others aren’t. So for example, it shows me that even when I’m feeling pretty good, I don’t think I’m very active or attentive. So now that I know that, I can try to do something about that and be more active or work on behaviors that help me be more attentive.
Make a little more sense?
Yes, it definitely helps but I can’t read them like you yet. I’ve looked at mine in view of your thoughts and it’s the same with me for the attentive and alert, i.e. they show up as the areas of concern in my positive emotions but won’t that be the meds doing that, leveling our our sharpness in the senses as well as our mood? If it’s not the meds isn’t that part of the distraction associated with ADD. I’m not sure. Luckily the bi-polar can kick in and enable me to hyper focus but then that can get too much on a role and be a bad thing as well but it can also be really useful. On the negative emotions, scared and afraid come up in red and dark orange but I am a big fat scardy cat. I’m not good without a fast plan and right now I’m in transition and trying to let things come together without forcing an outcome but to tell the truth I’m quite scared so that makes sense but might also indicate that I need to do some work on my intuition and strengthen trusting that I will make the right choices for myself. It’s interesting. I’m going back to have another look.
I think it can take awhile to get a sense of what to read into some of these representations, especially if you’re not used to it. I’ve spent a lot of time looking at and thinking about data displays, so I’m pretty good at picking out the interesting details. The harder part for me is figuring out what to do about it when I do identify a trend!
I agree with Maree. I really like Moodscope a) for Jon’s messages, b) for the standard mood graph but I really don’t get anything out of the triggergrammes or the affectogrammes. I know from teaching that different people need information displayed in different ways in order to understand them.
Whilst I am really pleased you get so much out of those two features, Dee Dee, I don’t know whether you have got Caroline’s ear more than we do but we have written rather more than once that we just don’t get these features at all displayed as they are … And the best that has happened is that we get referred to your blog … Clearly you are their ‘top of the class girl’ and the rest of us despite being post grad qualified must be the thickos at the bottom.
I would gladly pay for membership or further support development financially abut not if the developments don’t speak to me.
HOpe you are listening and can influence. You can probably tell I am going from interested to being rather frustrated …and one thing I know for sure is that frustration IS one of my negative triggers …and I needed neither the affectogrammes or triggergrammes to tell me that.
The best of luck with the dissertion … It really is exhausting writing up and editing. When do you have to defend it?
Ciao
Fionna
I can certainly sympathize! The fact of the matter is, this type of analysis and interpretation is something that I’ve spent quite a lot of time with for the last 7 years. And I do have a BS in maths – but I might add that study maths only provides analytical reasoning skills, and not any sort of experience in data interpretation beyond simple graphs!
I think it’s easier to get a better grip on the triggergrams as a starting point, since there are many word/tag cloud types of infographics available online and you can more directly control what appears by the way you write the notes. The simple frequency with which certain things pop up helps me think about what those themes have to do with what I’ve experienced on a good or bad day, and being able to click through to the original comments lets me verify that the connections I’m making are sensible.
I basically just ask myself, “why did that show up there?” Some of the words make little sense to me compared to others. For example, I can see that I was in a good mood when I turned in my dissertation manuscript, so that makes a lot of sense but isn’t something I need to concern myself with in terms of thinking about how to manage moods in the longer run. I’m not going to turn that manuscript in more than once! However, when I see that headaches and medication changes go together and make me feel awful, I can then plan to take it easier (or know that it’s going to be a less productive time) when I have to change medications, because I’ll probably get a headache. For me, interpreting the triggergram is a bit of a creative process, thinking about how things are related.
The affectogram is clearly a more complicated matter, though. It’s trend data like the line graph, but more detailed, and that is very different from just highlights. The problem with presenting each of the 20 Moodscope items in a line graph is that it would turn into a ball of spaghetti very quickly, and you wouldn’t be able to see any patterns at all. I used to make the same sort of heatmap graphics for “household name” clients quite some time ago, so it’s fairly intuitive to me despite being very differently presented.
I think one thing that would make it easier to understand this graphic would be if you could compare two time ranges, because it might help you see differences that you could then connect to experience. Being able to draw comparisons is very helpful for interpretation. That’s one reason I made these posts – I can’t see anyone else’s data, but by looking at my graphs and interpretations, it might help you see similar or different patterns in your own.
It’s the sort of thing that is easier to learn from example and through comparison, in my opinion. The dangerous part with doing this for mood data is that comparing oneself to others can be lousy for self-esteem, and it’s hard to make a fair comparison in the first place due to individual differences. This is why I think it would be more helpful to be able to compare your own data at different time periods.
Maybe this would be something where you might have an easier time if you could pick up to 3 or 4 of the 20 items and put them on a line graph together? That way you could look at some comparative trends without being overwhelmed with too much information – does that seem like it would be helpful?
My dissertation defense is in six days – thanks for asking! I’m now working on honing the defense slides. It’s almost comical to try to fit 2 years of work into a 20-minute presentation, but that’s the monumental task at hand.
Hi Deedee
Ouch .. Only 6 days…wishing you lots of Irish good luck.
I’ve been giving some though as to how the data could be presented… I’m no mathematician but have studied research methods and some stats…. And data interpretation and analysis … And presentation was a central part of my work (managing one of the NHL national screening programmes … So we were dealing with data on a very large scale. That is why I think tht if I am struggling with the data so will others.
I agree that part of the problem is that there is too much information in the two features … Particularly the word clouds… Whilst I can see tht one way of dealing with it is to limit one’s own use of adjectives…that would have the disadvantage of removing important nuances from one’s own record.
I might prefer to have my top 10 adjectives and top 10 names in the positives and negatives … But I can see disadvantages in tht as well. Perhaps when your thesis is defended we might have more time to discuss… What’s the hypothesis that you were investigating in your research?
Fionna
I think one solution to helping with the word clouds is to allow selective (temporary) removal of terms. This is a feature I’ve seen on Wordle that I really liked. It removed irrelevant words upon my selection and then re-weighted the others – that would be one way to address the “top ten” sort of preference that you mention, without necessarily removing the important nuances.
Stemming would also help, by combining words like “tired” and “tiredness” or “outing” and “outings”, so that you see more clearly what is meaningful. You can control that to some degree by being very consistent with what you enter, but that is often hard for people to do since it requires quite a bit more cognitive effort than most of us can muster on a lousy day. That’s where a more tagging-like implementation or option could be really useful – I’d love to see word clouds for both notes AND tags. With tags, you can get suggestions from your most-used terms, which makes it substantially easier to be consistent about how you’re describing things.
No hypothesis in my research; it’s thoroughly qualitative social science in contrast to most of my prior experience, which is with large-scale secondary data analysis! But it was the most appropriate approach for the questions I wanted to address, so the study oriented toward descriptive empirically-grounded theory development about a complex social phenomenon.
I’d love to say more (because I really can and will talk all day about it…) but the topic is so very distinctive as to be immediately de-anonymizing! While everyone is supposed to do something unique with their dissertation research, mine is focused on a phenomenon that hasn’t gotten much attention yet, though that’s swiftly changing – which conveniently puts me a bit ahead of the curve, so to speak. It’s the sort of thing that I can easily talk about with a seatmate on a plane, and they say, “Wow, that’s really cool! I see why that would be important and I’m going to go check it out.”
I think what we should do is organize a Skype call with face so that DeeDee can take us through a graph step by step when we have the graphs right in front of our screens. i.e. we all look at the same graph and learn how to interpret it •*¨*•.¸¸♥
Happy to participate in a Skype call … Although I won’t be able to view my affectogrammes or triggergrammes as I decided they did not ‘speak’ to me at all and I didn’t seem able to engage with those in Moodscope to discuss this meaningfully. I am afraid that I strongly disagree with Deedee that these are in the least bit intuitive..they are too complicated in their display to do that… So if we do have a calll (and I would love to hear about Deedee’s presentation … And call her Dr Deedee…) …I would very much like to discuss how these data might better be presented . I hope that Jon and Caroline will understand that users will be happy to pay for the service and support further development providing the data ‘speaks’ more clearly and colourfully. Deedee may regret being pulled into this given how much else she has on her plate! Cheers from England. Fionna
I’m afraid I can’t be pulled in any further than I already am!
Lovely as further discussion might be, I have more than I can manage already. This is my “for fun” stuff, and teaching about data is (to put it rather bluntly) something I get paid for. I mix work and pleasure almost by profession, but adequately balancing my load requires trying to keep the fun stuff fun and the work stuff work. And knowing when to say no, which is something I’m terrible at. So you’ll have to forgive me, but it’s for my own good! I’m sure you can understand.
No probs. I used to find that I kept on getting dragged into work related stuff & it ruined my precious leisure time…good for you saying no!
HA! I just a link to this post from MOODSCOPE themselves. You are a celebrity, aren’t you!? Thanks for talking to us lowly folks. Big fan, Dee dee.
I have found the affectogram very useful. one reason I am prepared to pay for it. the triggergrams are kind of how I expect.
the daily email is excellent. I am very glad to have found moodscope.
of course it is from an email that I found out about you!!
I find them both useful and fairly easy to understand as well – not everyone does, but the word clouds in particular do seem straightforward to “read”. Interpretation is another matter entirely!
the word cloud thing does make me realise that some things that are assumed to be triggers – by me and therapist I recently saw – are not always trggers but can be. I’m thinking in particular of physical pain.
The affectogram was just brilliant. so much more help than the activity and mood rating 1-10 that I was doing for CBT, which just confused me because there were several moods and they would want different scores. On Moodscope it became so clear where problems are – and made it clear how good days were achieved – they were achieved basically by acheivement! And the little graph at the bottom – the blue bits – can be very interesting
but then yes there is what to do about the trends…
Pingback: Weekly List Bookmarks (weekly) | Eccentric Eclectica @ ToddSuomela.com
Hi, I have to say I like the ‘word clouds’ but it took me a couple of weeks to work it out. I realised that putting ‘lovely weather’ and ‘horrible weather’ only brought up the word ‘weather’ and put it on both positive and negative which confused me at first, till I really realised that it takes from your 10 most recent positive scores and your 10 most recent negative scores and so ‘weather’ was popping up in both as I had been writing righting about it in my comment for the day. its as easy as that! you do need to try and use the same lingo and not become a thesaurus as this wont help track your moods as easily.
I still have to get to grips with the cards graph using the heat colours is a good idea but as dee dee suggests I think it would be better set out differently.
I’d also love to be able to print out my months worth of mood from all 3 charts (or download) so that I could keep a record of them to refer back to as and when i need them.