Analyze your zombie walk
Imagine a zombie costume.
You probably see – pale, bloody makeup, ripped clothes, dead-eyed stare. All of that preparation, however, would not be convincing if the zombie were to just move around normally.
In fact “Walking Dead” special effects artist/director Greg Nicotero teaches a school on how to lumber around like a zombie for extras on the show.
I had the same question as this interviewer, so I quote directly:
“MJ: How much acting skill do you really need to stagger around and snarl?
GN: More than you would think.” 
He goes on to say:
“We try to avoid the Frankenstein look where their arms are outstretched. Or someone will drag their leg and look like the Mummy or another person has their hands in claws next to their face like The Wolfman. All these people are going for the classic movie cliches, which we try to break.”
Instead, Nicotero asks his zombies to look kind of drunk. Slightly slow, with delayed reactions. 
Where did the stiff outstretched arms cliché come from? If you have read Ashley J’s post from last year, you will know that real (Haitian) zombies were probably created with TTX plus a combination of drugs. However this mixture produces catatonia rather than stiffness.
So why are zombies rigid? Perhaps like newly dead corpses, the undead are experiencing rigor mortis. This is muscle contraction resulting from biochemical degradation within the muscles.
Since (most) zombies are also fictional, I won’t provide an answer to this question. Instead lets just agree that zombies probably don’t have an intact brain, and segue into what that could mean for the way they move. Or conversely, what being rigid or drunk-looking means about a zombie costume wearer’s costume brain.
The “Frankenstein” zombie
Sir Charles Scott Sherington, Nobel Prize winner in 1932, determined several types of “abnormal posturing” by studying lesioned cats and monkeys. Posturing in this case means involuntary extension or flexion in the limbs, and indicates severe brain damage when seen in human patients.
The most severe example is damage to the brain called decerebration, as in removal of the cerebrum. This is equivalent to damage in the brainstem at the level of the pons, because it cuts off input from everything above it. Decerebration results in the elbows, knees, and tail becoming rigid and extended, while the wrists and ankles flex.
Sherrington called this response “peculiar” and named it “decerebrate rigidity.” He contrasted this rigidity with lesions below the brainstem, in which the animal’s extremities are limp and slightly flexed – which means that the brainstem provides input to the extensor muscles causing them to extend.
If removal of the cerebrum causes flexion and rigidity, this means that its normal action is to inhibit this flexion. Actually, it isn’t as simple. Damage further up the brain results in “decorticate posturing” (as in removal of the cortex). In decorticate posturing, the legs are extended, but the elbows are flexed, and hands are clenched.
It turns out that the key difference between the two types is whether an area called the red nucleus is present. When the cortex is gone, the red nucleus is no longer inhibited, so it becomes overactive. As a result, the path going from the red nucleus is overactive, and it drives the flexor muscles in just the upper limbs. It turns out that it is strong enough to override the brainstem drive to extend the limbs.
What does this mean for a rigid, arm-extended zombie? First of all, you may wonder if it should be able to move at all. Actually, this is the research has laid the groundwork for the idea that movement happens by reciprocal excitation and inhibition. The spinal cord has its own sets of motor loops that consist of sets of neurons that alternately drive the muscles on either side.
Here is a video of a decerebate cat walking on a treadmill. https://www.youtube.com/watch?v=wPiLLplofYw
Without a cortex, the cat moves because when the motor neurons on one side are active, they inhibit the ones on the other side. At some point, they become less active, and then switch, so that now the other side is active, while the first side is inhibited. The result is an automated walking pattern, that can be activated and changed slightly by inputs from the brain or sensory receptors.
Practically, the rigid zombie should be somewhat of an automaton. It can have extended or flexed arms, or one of each, and should magically be able to step forward.
What about the slowed and uncoordinated zombie type?
The slightly drunk looking zombie.
I’ll refer you immediately to this very creative blog post. The author suggests that the more sophisticated zombie movement sounds most like damage to the cerebellum. This condition is called cerebellar ataxia and results in poor movement coordination, a wide and staggering gait, as well as slurred speech.
Cerebellar lesions were studied in detail during WW1 by Sir Gordon Holmes, a neurologist who described the associated motor disturbances in wounded soldiers. He saw that reflexes were still present in his cerebellar patients, despite apparent lack of muscle tone. He concluded that the cerebellum coordinates voluntary movement, to set the range and precision of movement. In fact, the cerebellum does mostly sent input to other motor centers rather than directly to the spinal cord. [6,7]
A cerebellar zombie, should have trouble with fine motor control, and perhaps timing its movements. If you really want to test your costumed friend, you can try a test developed by Holmes and his collaborator Stewart, known as the Holmes Rebound phenomenon. Have the zombie pull on your hand, and then suddenly slip your hand out. Normally, a person can adjust their movement, and will jerk their hand in the opposite direction. A convincing cerebellar zombie will be unable to change direction of motion, and will continue to pull their hand back. 
Other zombie types?
Hopefully, you should now be able to recognize the basic zombie walker types as you zombie watch this halloween. Of course, since they are already dead, anyimaginable combination of brain damage is possible and even plausible in a zombie.
I will to refer you just one more example to help you classify the halloween walkers. Here is a video of example zombie gaits interpreted as muscle damage and pain. No neural damage at all!
 The integrative action of the nervous system, Sir Charles Sherrington.
 “Control of locomotion in the decerebrate cat.” Whelan PJ 1996 Prog Neurobiol
 “Sir Gordon Holmes” C S Breathnach Med History 1974
 Historical note J M S Pearce J Neurol Neurosurg Psychiatry 2004;75:1502-1503 doi:10.1136/jnnp.2003.016170
 Stanford school of medicine 25, Cerebellar exam. http://stanfordmedicine25.stanford.edu/the25/cerebellar.html