myboys.me/Shutterstock

Several pieces of legislation across the world are coming into effect this year to tackle harms experienced online, such as the UK’s Online Safety Act and Australia’s Online Safety Act. There are also new standards, regulations, acts and laws related to digital products (including smart devices such as voice assistants, virtual headsets) and services such as social media platforms.

Of the many harms these types of legislation are designed to address, “psychological harm”, “mental distress” or similar terms are commonly included.

Unfortunately, when psychological harm and the like are referred to, there is typically no detailed corresponding definition of them. But while we might have an intuitive understanding of what psychological harm is, we still need precision on what it means in law. This means evidencing what it is, agreeing on how to measure it and designing the best methods to tackle it.

How do we do this? An obvious place to look is psychological science.

The origin story

The earliest reference to psychological harm was made in the 1940s. Back then, it was about the destabilising impact of war propaganda and the use of psychology to subvert people’s understanding of reality. Psychological harm was a broad term, which also applied to those witnessing the horrors of war on the front line.

In the 1950s and 1960s, psychological harm was more associated with advertising tactics that aggressively exploit people’s emotions and insecurities.

Fast forward to the early 2000s, and tools for assessing psychological harm emerged alongside clinical assessments of mental health disorders. For instance, research on abuses experienced online, such as cyberbullying and cyberstalking, documented several psychological impacts. These ranged from withdrawal from social groups, self-doubt and reduced self-esteem to mental health disorders such as depression, anxiety and PTSD.

More terms entered into clinical and forensic lexicons, such as “psychological distress”, “psychological damage” and “psychological injury”. All of them concern some form of mental adverse experience which may happen immediately or as a delayed reaction to traumatic events.

Where we are now

In reviewing the 80 years’ worth of work in clinical, forensic and cognitive psychology, here is what I see as the major issues concerning psychological harm.

There is no agreement as to where to draw the boundary between psychological harm or related concepts and mental disorders outlined in the diagnostic manual called DSM-5-TR (such as depression, anxiety or personality disorders).

There is also no standardised measure of psychological harm or psychological distress or damage. For instance, if we just take social media, there are different metrics that vary even on how they measure negative mental experiences on Tiktok, Instagram and Threads, Facebook, Youtube and Weibo.

Cyberbullying can cause major harm.
Ground Picture/Shutterstock

Why does this matter? Take for example cyberbullying. There are 17 tools in existence to measure psychological harm. And because the tools don’t all align, we don’t have an accurate picture of rates of psychological harm. Some tools are too narrow in scope – they fail to include severe cases that require psychiatric treatment. And other assessments are too broad – failing to exclude those that are malingering.

What’s more, how we perceive and experience adverse events, which can be very serious and debilitating, vary – they are subjective in nature. Research in clinical and forensic psychological recognises this. These disciplines have spent time establishing standards of assessment when supporting legal decisions for ensuring appropriate punitive measures when we face terrible situations.

Three practical suggestions

For legislation to do the job of guarding against psychological harm from serious adverse experiences online and through digital technologies, forensic psychology offers a path forward.

The first thing is to have an agreed definition. For example, in 2025, the psychologist Amanda Heath proposed a viable general-purpose definition as “a sustained drop in stable functioning, negatively impacting wellbeing”.

This works in the same way as legal requirements for defining physical harm, which needs a baseline of functioning to show how an injurious event causes a change to it. The severity of the damage varies, based on, say the length of recovery (such as a week, a month, a year, never). In the same way, the length of recovery from exposure to illegal content online would indicate the severity of the psychological harm experienced.

Second, there should to be a process for demonstrating causality between a particular adverse event online and the harm itself. So far, there doesn’t appear to be any set criteria laid out in online safety or harm acts for establishing causality.

Again, legislators could learn from forensic research, which outlines two levels in psychological injury cases that establish causality – psychologically and legally. Forensic psychologists weigh the evidence for the relative ratio of pre-existing and event or post-event factors to determine causality using something called counterfactual analysis.

For example, sometimes people have pre-existing injuries, vulnerabilities, or psychopathologies. So in such cases there needs to be a baseline, where the evidence shows how an indiviudal’s conditions have been exacerbated by experiencing an injurious event. For example, if we applied this analysis to psychological harm experienced online it would work like this. Forensic psychologists would weight the evidence to determine that, in the absence of seeing the illegal content, an individual would not have experienced PTSD to the same extent that they are experiencing it currently.

Finally, there need to be standards for the evidence used to show causality between a particular adverse event online and the harm itself, which we don’t yet see in current online safety or harm acts.

In forensic psychology, on the other hand, the legal standards of evidence are high, requiring independent corroboration of psychological impacts. This is where psychiatric assessment tools of PTSD, depression and anxiety are used along with other sources of evidence. Physical outcomes (such as neurological damage) and behavioural outcomes (such as substance abuse, self-harm) are also required.

To serve the public, the law needs to improve. This can’t be achieved without a fleshed out definition of psychological harm, tools of assessment and a framework that traces a causal path from the injurious content to the harm it is considered to have caused.

Magda Osman receives funding from ESRC, EPSRC, Research England, UKRI Innovate UK, Wellcome Trust, Turing Institute, Food Standards Agency, DFG, British Academy, DSTL, Counterterrorism Policing.