The study, conducted by Emsense on behalf of Gamasutra, used bio-sensory headsets to measure the engagement of gamers while playing titles such as Half-Life 2, Halo 2, Gears of War and Battlefield 2142.
"Close combat was the most reliable method of creating engagement, adrenaline, reward, and all the emotions that make shooters so much fun," wrote Emsense's Tim Hong. "Certainly, this is nothing new to the genre, but the next-gen games that excelled in this area were exceptionally strong at creating high-paced close combat frequently."
In one graph recording a player's feedback during a Gears of War close-quarters attack, the "engagement" meter spiked following the landing of a chainsaw, indicating that the visceral nature of close combat is in itself a highly rewarding act.
"Gears of War players recorded high emotional reward for the spray of enemy blood after they succeeded," said Hong. "Of course, we can't forget the ubiquitous Half-Life 2 crowbar, the only weapon players initially have for fighting."
Outside of statistics collected from each individual game (see chart above), the study also compared and contrasted similar sequences amongst competing titles. One particularly interesting examination involved turret sequences in both Resistance and Halo 2, wherein Resistance failed to adequately engage players due to a lack of close combat danger.
"The failure lies in how protected the players are," wrote Hong. "In Resistance, one of the players' experiences with turrets in the first 90 minutes is from within a huge tank. In Halo 2, players utilize small, unprotected turrets that nearly ensure that they will be harmed, if not killed, if they remain on the turret for long."
The study also reinforced the "rollercoaster" theory of game pacing, where lulls in the action enhance the combat that follows--a strategy adopted most prominently by Epic with Gears of War.
"It seems counterintuitive, but the most intense points of engagement in the titles in our study were often the result of calm moments," reads the study. "Downtime, a period of lower engagement, is not always bad. Periodic but brief lulls in action allow for more intense action sequences and stronger reactions to climactic final battles."
correlation not causation?
except for the part where EEG "intensity" is just an excuse to make a series of qualitative observations
No, the time course is pretty clearly laid out and peaks after the start of close-quarters combat, which pretty firmly isolates melee combat as causative to an increase in their measure of "engagement."