It Takes Two to Benefit Cortisol and Testosterone levels

I’ve recently taken up learning tango dancing as a new hobby. I’ve been doing it for several months now and have to say that it’s quite enjoyable. I mentioned to someone in conversation how when I dance, I feel very elated afterwards, and they remarked that there’s probably quite the testosterone response from tangoing. Not being one to let a good opportunity for over-examination go to waste, I did a quite good search for articles on tango and hormone release. Turns out that one exists. Here’s a link to an abstract.

The study tested levels of both salivary cortisol and salivary testosterone before and after dance. The study was set up as a 2×2 study looking at dancing with and without a partner and with and without music to compare the effects of both variables alone and in combination with eachother.


Results showed that dancing with a partner showed the greatest drop in cortisol levels, however all test situations showed drops in cortisol. The drop from tango dancing with both music and partner showed nearly double the drop in cortisol on average compared to other test situation. The presence of music seemed to have a large effect on this drop, as the least significant changes were seen in cases without music.


Effects on testosterone seemed to be largely dependent on presence of a partner. Change with a partner was M=2.66 pg/ml, SD=4.45; and without a partner was M=–0.260 pg/ml, SD=4.23. So testosterone levels actually decreased slightly (on average) when dancing without a partner. Interestingly though, the study designers chose to repeat the situation of the first week (with partner and music) on the fifth week to see if perhaps the nerves or novelty of the first week could have caused a difference in testosterone levels. As it turns out, it had a huge effect. Session 1 showed an increase  of 4.06 pg/ml, SD=5.77; yet when they returned to the same conditions in session 5, they saw -1.17 pg/ml; SD=7.19 pg/ml.

From my experience, I would hypothesize that this change was a result of novelty. Dancing with a new partner can be extremely exhilarating (Coolidge Effect fallout anyone?). Although all participants had over one year of experience dancing and all but 4 enrolled with partners with whom they were partnered the entire time, the study does not mention the frequency or recency with which they couples had danced together previously. I suspect that the regularity of the study caused some of the novelty to wear off by the 5th session. One interesting follow-up for this would be to have one group of couples which would switch who they danced with and a control group which remained with the same partners and measure the effects on testosterone over the sessions.


I’m going to keep dancing because I enjoy it, but it is nice to know that I’m helping to reduce my cortisol and increase (maybe) my testosterone levels. I’d be very curious to see a larger study done with more participants and some additional factors. This one only had 22 participants, and while results were tested for statistical significance, it’d be nice to have more data.

A Swift Kick in the Joulies

Edit: As noted in the comments, I’m testing these under a situation which is not the recommended method from the Joulies company. I am, however, testing them in the real-world situation in which I would most likely use the Joulies. I wanted to add this disclaimer to be fair to the guys struggling to turn this into a company. I stand by my results, but leave you to make the decision about how well the Joulies would meet your needs for the product.

Abstract: I recently received my set of Coffee Joulies ( after putting some money into their Kickstarter project. Excited to review and test them out, I came up with a simple experiment. I tested the Joulies vs. an empty control mug with just hot water vs. a mug with rocks of roughly equivalent size to the Joulies. Both the Joulies and the rocks dropped roughly 30 degrees in temperature immediately and then lost steadily at the same rate. The cup with just water didn’t see the immediate drop, but lost temperature steadily. I saw no significant or noticeable difference between the rocks and the Joulies. Given that the Joulies cost $50 and the rocks cost me walking out into the front yard, I’m skeptical about the value of the Joulies.


After finding  3 mugs of identical size and shape, I went out in the yard to find a handful of rocks which closely matched the volume of the Joulies. Following this, I set all 3 mugs up on the same wooden coffee table surface. I left the mugs, rocks, and Joulies overnight to allow them to equalize to room temperature.

The following day I boiled enough water to fill all three mugs in an electric tea pot. Once finished, I began the testing. Filling the first of the three mugs (the empty one). I started my stopwatch. I measured the temperature immediately using a digital meat thermometer. My method for obtaining temperatures was thus: place the thermometer in the water in the corner opposite the handle (all mugs were oriented the same way), then watch the temperature until it flattened out and began dropping. I measured the temperature as the maximum before the drop. Typically it only wavered by a degree or so.

Thirty seconds later I poured the next mug (with the rocks) and repeated the measurement for it, I then did the same with the 3rd mug (with the Joulies). By doing it thus, I was able to measure all of the mugs at the same elapsed times. I then continued this process at intervals until reaching 1 hour.


As usual, a graph is worth a thousand words. I’ve broken them up into two different graphs to prevent distortion from having changed from measurements every 2 minutes to every 4 minutes at the 20 minutes mark.

Minutes 0-20
Results from Minutes 0-20

Minutes 20-60

No that we can see the behavior of the 3 mugs and their contents, let’s examine some claims from the Joulie website.
1) “Their polished stainless steel shells are full of a very special phase change material (an ingredient in food) that melts at 140°F. When you put them in your coffee this PCM begins melting, absorbing a LOT of heat in the process and cooling your coffee down much faster than normal.”
I can’t argue with this, they did indeed rapidly cool the hot water down. However, the rocks and the Joulies had roughly the same effect in this regard.
2) “When your coffee reaches 140°F (the perfect drinking temperature) the molten PCM [Phase Change Material] begins solidifying again, releasing all that energy back into your coffee to keep it at a comfortable and delicious drinking temperature. The more heat you feed your Joulies, the longer they’ll keep your coffee warm.”
The Joulies behaved almost identically to the rocks and as such do not seem to have had any sort of magical technology. In fact, after the initial drop, the rocks maintained a higher temperature for a longer period of time. Also worth noting, the slope of the rocks and Joulie lines were very similar, showing a large initial drop and then a steady decline.

Obviously, this is a fairly limited experiment and it would be helpful to perform with various other size containers and materials. However, I do think that it is enough to show that the Joulies are not some sort of magical material. While there may indeed be some sort of special materials inside, the effects they had were not enough to be noticeable in a real life situation, and therefore do not justify the cost of the Joulies over simply buying a more massive mug.

Edit: A friend just sent me this similar post on testing and reviewing of Joulies by Marco Arment

Edit 1/15/12: Here’s another interesting article on the subject.