Surprise! Fighting Familiarity to Maintain Player Enthusiasm

Foreshadowing – Sometimes teasing new content is better than giving it away. Creating anticipation (like nearly everything in game design) is an art form, but the principles are simple. You want to show or talk about something that players desire or are intrigued to discover more about, then use that desire to drive them forward, asking them to overcome bigger and bigger hurdles while constantly reminding them what the reward for the work is. Zelda games often do this well, but an outstanding example can be found in original Unreal from 1998.
If my memory doesn’t lie, the game has a level that features a 6 barrelled rocket launcher positioned in the middle of a large chamber, on a top of a stone pillar, far out of player’s reach. You are then off to work, to earn the weapon. You navigate a net of corridors and rooms around the location, fighting enemies and slowly activating various mechanisms. After all the hard work is done, you can walk across a newly opened pathway, finally grabbing the weapon, which is one of the best in the series. 
Great games are capable of turning what’s otherwise hard and mundane work into piles of fun. The point is, tease your players occasionally to maintain their interest. The presentation of your reward is as important as the prize itself (that’s why sensual lingerie works so well).

Mystery – This is a huge topic that I can’t tackle fully in this post. Nevertheless, mystery surrounding the game world, objects, back story and main plot can be utilised for extra engagement, excitement and creating surprises (once mystery is unveiled). Often it’s good to be a little vague about your game world, it makes people ask questions and seek for answers, pulling them in.
Dark Souls does it perfectly with the story of the world and NPCs which is entirely optional. The game provides information, but only to those who seek it out. There are lots of clues hidden in item descriptions, extra NPC dialogues etc. Scraps of backstory which take time and effort to reconnect. Even then, there are still so many questions left to answer (and that’s what leaves you wanting more).

Random Surprises – Think about systems that can yield surprising outcomes based on a range of parameters and probability, or about preparing surprises to be triggered in their own time. Weird chains of events, randomised loot drops, unexpected encounters and unpredictable outcomes to interactions between elements of the game world (often AI doing things we didn’t realise they were capable of), are all very personal and exciting. If you are building some features around probability and parameterisation, there is a chance you can use them in a good way (perhaps you may even want to cheat a bit and plant some ‘good seeds’ into the system yourself). Borderlands loot drops are a straightforward example, but there are more vague ones too. I felt that the random encounters in Fallout 3 were more thrilling and memorable than the rest of the game world. So was the Mysterious Stranger, a by-product of a weird perk that would occasionally summon a man with a coat and a magnum to help you out (accompanied by a cool guitar theme).
Eternal Darkness’ famous sanity effects are all semi-randomly triggered depending on the player’s state of mind and location. They are always surprising and very often jaw-dropping and frightening.

Powerful Twist – Standard and a bit underappreciated due to too many bad examples, but good story twists go a long way towards creating memorable experiences. I don’t feel competent enough to talk too much on the subject, but if you ask me, I think a good twist is one that you don’t see coming despite having ‘all the clues’ laid out in front of you and despite it being believable and making sense. It’s a ‘eureka’ and ‘of course!’ moment, you’ll find a better source than me to talk about it in depth. Some good examples might include Braid (ending), Portal (middle) and Bioshock (ending) – they all displayed clever and surprising story twists that elevated good games towards greatness.

Confounding Expectations – It’s good to pull out an interesting feature or a great moment out of nowhere, breaking pacing conventions and giving players a new toy before they subconsciously expected to see it. Remember, people notice patterns everywhere, if you give out weapons every 3 levels, surprise them and give them a new gun just as they got familiar with the previous one. Ni No Kuni introduces new features as late as 40 hours in! The key is to reinforce intriguing uncertainty. Another example would come from Mario & Luigi: Bowser’s Inside Story, a game that’s already big, full of great little surprises and rich in content. However, I’ve never expected that half way through a 40 hour long game Bowser would get crushed by a FLYING CASTLE, which then caused him to go frenzy, growing to a size of Godzilla. The game then asks you to turn your DS around into portrait mode and delve into a battle of giants with the whole island as your battleground. Entire new combat system with new moves and new character to control. A jaw dropping moment that’s unsurprisingly the most memorable of them all.

Expanding Depth – This has a lot in common with ‘confounding expectations’, but this point focuses on adding depth to systems that players perceived as final, complete (‘cracked’ and mapped in their memory). It all relies not as much on adding more stuff, but on obscuring certain elements before you want them to come into play. Not only will it allow players to focus on what they have at hand, but also surprise them pleasantly when there’s more to discover in familiar areas.
To bring you some light examples, imagine adding a 4^th character slot to your RPG party of 3, half-way throughout the game, changing team composition and possible tactics without actually adding more content. Or in Borderlands, giving people a way to ‘mix guns', allowing them to gamble their loot in a hope of getting a good combo of stats or upgrade the alchemy to allow mixing more ingredients for more potent and cumulative effects. So building on what you already have, adding ‘extra depth’ with new strategies and possibilities to explore, in places that seemed finished.

Perfect Size/Content Ratio – This stretches across game design to content planning and production. Very often games can sacrifice features in order to fit into a given budget or time frame. The first thing that often gets cut are the ‘nice to have’ additions that aren’t core to the experience (at least in isolation). If this is done excessively without readjusting the ‘length’ of the game, we might have a problem of density. Too few elements will be spread too far in between, with hours of ‘filler’ content just to bridge the gaps in the story and ‘check’ levels off the list. All these ‘nice-to-haves’ are some of the main reasons behind good indie games taking so damn long to finish. You do not want to create a long game that loses players due to boredom half-way through, better to plan game length accordingly to feature-set. Of course you cannot go into extremes and try charging $60 for a two hour game; price has to match the scope too. Gone Home and Brothers: A Tale of Two Sons are both great examples of very well balanced games. They are short, but I’m not sure they would work so well if they had been longer. The stories would lose their emotional punch when tackled across many days, and in case of Brothers, stagnation could set in when it comes to its innovative puzzles and controls.
Don’t outstay your welcome! It’s better to have a short game that’s full of little touches and hidden gems, than a long but fairly repetitive and predictable one.

Bending The Rules – Do this carefully and consciously as it may have unwanted consequences in form of confusion and frustration. This has to be properly communicated to the player and meaningful. In such cases, changing the well-established rules to deliver something fresh can be a good thing. An example of bending rules would be infusing your players with a fire spirit, making them heal when standing in fire instead of losing health. Changing the way the game is played with a small modification to the rules. Pacman bends the rules and reverses the roles of hunter and pray whenever you pick up a power pill.

Our Little Secret – Sometimes the game is not as much defined by its “main completion path” but by everything that’s hidden from plain view. Even the elements that are hard to discover could go viral and re-ignite interest in the game or build a strong ‘promise’ that there are still things to discover. Easter-eggs/small secrets for players to discover often create very memorable moments and also encourage the finders to share their discovery with others (you won’t share ‘oh I killed that boss!’ as an achievement, everyone who played the game did it, but a secret on the other hand is worthy attention!) Portal’s secret rooms and Rattman’s den are exciting for many, with countless fan movies and comics surrounding a completely optional part of the story. Sometimes less is more, let players fill in the gaps themselves, their imagination is often better than your cutscenes.

The Small Stuff – This is a difficult and perhaps a generic one. However, there are certain parts of games that we live to remember forever, yet they are neither big, nor secret, nor affect gameplay. It’s all about those silly moments that make the game less ‘mechanic’, more human, often funny. The messages on walls in Left 4 Dead’s safe rooms, the (often hilarious) mails found on computers in Deus Ex: Human Revolution or the ridiculous dialogues between the guards.
All these tiny things add up to a more personal & memorable experience.
Looking back at Mass Effect 2, it was a quite forgettable experience for me; the game blends in with the whole series into a single clump of memories. That said, I’ll never forget Mordin’s song, a moment that defined the game and made me genuinely happy.