Why we need an online version of the Trier Social Stress Test

The past year has been quite a challenge for most of us who perform experimental studies with human subjects in psychoneuroendocrinology. In response to the SARS-CoV-2 pandemic, most universities and other research institutions still prohibit in-person.

Data collection to avoid further spread of the virus. So, what can we do? Stay in limbo and wait for better times to bring back our old lives? Rather than to freeze and lament about the injustice of life, why not accept the COVID-19 challenge and try out new things that may change the way we do research even after this pandemic is over?

Megan Gunnar and colleagues from the University of Minnesota opted for creative thinking and action. Taking advantage of the ever-increasing speed of internet connections, the pervasive use of mobile phones, and the availability of online communication platforms that allow for seamless video and audio streaming, they now show that acute endocrine and psychological stress responses can be induced remotely – in a Zoom™ call. Not that students, teachers, or (executive) employees would not have felt being stressed by and during zillions of online ‘meetings’ and seminars we all have attended in recent months. But the mere feeling of being stressed does not necessarily translate into a significant release of stress hormones. Especially the hypothalamus-pituitary-adrenal (HPA) axis is quite picky with regard to the appropriate stimuli for initiating a secretory episode. Only under novelty, unpredictability, and/or uncontrollability of adverse stimulation will we respond with acute stress-induced increases of ACTH and cortisol levels. While physical stressors can be quite potent in this respect, too, threat to the social self is especially powerful to startle the HPA axis.

Jeopardizing the social self is the key psychological element in the most frequently used stress protocol to date. Almost three decades ago already, the Trier Social Stress Test (TSST; Kirschbaum et al., 1993) has been shown to elicit robust biological and psychological stress responses, including the HPA, under laboratory conditions. Although virtual reality versions of the TSST have been shown to be quite useful (e.g., Kelly et al., 2007), this and other variations of the protocol still require personal contact in the lab.

The paper by Gunnar et al. (2020) in the present issue of Psychoneuroendocrinology shows that we can do without: Even in the absence of real person-to-person contact, our ego and social self can be threatened quite effectively. They describe an online version of the TSST (dubbed TSST-OL), which they evaluated in a group of teenagers. Cortisol, alpha amylase, and subjective responses to this TSST-OL were found to be comparable in magnitude and temporal pattern to those observed in response to the ‘classical’ TSST. Two out of three teenagers displayed a cortisol response, which is in synch with other reports, but still a bit short of the response rate one can achieve. Giving the test person a glass of cranberry juice (or other form of glucose load) 20 min before the stress exposure, will further raise the responder rate to >85%. In contrast, fasted subjects with low blood sugar levels tend to be HPA non-responders (Kirschbaum et al., 1997). Good news also for those who study adult populations: The TSST-OL works just as well with college students and older adults according to a pilot study conducted in my lab (unpublished data).

Gunnar et al. should be congratulated for this important methodological advancement. The TSST-OL is a resource saver: Travel time to the lab for study participants and researchers is zero, and no precious testing space in our research institutions is needed. Independent of current or future lockdowns with closed lab doors, everyone can now conduct a study with acute stress induction. Thanks to them for also sharing the details of their TSST-OL protocol through the Open Science Framework (https://osf.io/aqg9u/).

Apart from enabling us to conduct contact-free stress experiments in a resource friendly way, there is yet another big advantage of the TSST-OL over the classical lab-based protocol. Voiced by another Minnesota-based researcher during a recent online-held convention, Mustafa al Absi hailed the new stress protocol because of its inclusiveness: We can now perform stress studies with individuals from far more diverse cultural, ethnical, and geographical backgrounds. Imagine to do a TSST study with subjects participating from basically all over the world!.

COVID-19 has changed our lives in many ways. Instead of real person-to-person contact we have learned that virtual meetings can be just as efficient in business and science. The TSST-OL is a powerful research tool for no-personal-contact stress induction, which is great. But just like in business negotiations, we will certainly miss a great deal of information about our study participants should we fully rely on internet-based interactions. Despite all this, the TSST-OL protocol is a true methodological gem for many PNE stress labs doomed to idle until the pandemic is over. And it will remain to be an important addition to the PNE toolbox well beyond the current crisis.