ASSAP: Paranormal Research
ASSAP: Paranormal Education
Privacy and cookie information ASSAP mailing list

Do ghosts emit electromagnetic fields?

Magnetic field produced by a washing machineThere is a widespread idea among paranormal researchers that ghosts emit an electromagnetic field and that their presence can, thus, be detected by EMF meters. A related idea is that ghosts may disturb the existing background (geo)magnetic field when they are present. Either way, 'unexplained' changes in EMF meter readings are often taken as a sign of ghostly or paranormal activity.

In spite of these claims, there appear to be no formal studies to support such ideas. Instead, there are a few anecdotal reports that EMF meters (see here for what they do) 'spike' during paranormal activity at haunted locations OR that haunted locations produce more variable EMF fields than non-haunted places. In both cases, it is difficult to trace any original, first-hand reports of these claimed connections and what reports are available are vague and lacking in crucial technical detail. So, why do some people claim that ghosts emit EM (electromagnetic) fields?

Asking the right questions

Do ghosts really emit, or disturb, electromagnetic fields? Before we can answer that question, there is a more important one to be answered. Are EMF meters even capable of demonstrating that ghosts emit EM fields? And before we can answer that question, we need to ask - do ghosts even cause hauntings?

So, firstly, do ghosts cause hauntings? The answer isn't as obvious as you might think! No one denies that hauntings exist. People undoubtedly experience odd goings-on at certain locations from time to time, but are ghosts really responsible for them? This may seem an odd question until you realise that apparitions are only witnessed in a minority of hauntings. What is more, there are few, if any, accounts of apparitions actually 'doing' any of the things we associate with haunting. They are not seen knocking on walls or tables, moving objects, whispering in corners, nor is their rare appearance usually accompanied by strange smells or a sudden feeling of cold. It is just assumed that ghosts are doing all these things associated with hauntings.

Thus, the whole idea that a ghost is responsible for, or even vital to, a haunting appears to be based not so much on evidence as popular culture. Looking purely at the evidence, apparitions may simply be one possible, nonessential, phenomenon that can appear during a haunting (see here for more on this debate)! Furthermore, investigations reveal that many ghost reports are produced by misperception or near-sleep experiences, both of which are subjective and so incapable of causing magnetic fluctuations.

It therefore seems to be an assumption too far to expect EMF meters to address the question of whether ghosts emit or disturb EM fields (for other popular assumptions that go beyond the evidence in ghost research, see here). Looking purely at the evidence it is therefore more meaningful to ask if EMF meters are capable of answering these two questions (based on anecdotal observations) instead:

  • are hauntings associated with elevated or more highly variable EM fields?
  • is observable paranormal activity associated with EM field spikes?

Why use EMF meters at all?

EMF meters are designed to measure mains-frequency magnetic and electric fields in buildings to see if they exceed certain levels*(see also here). It is, therefore, difficult to see why such a meter should detect ghosts, given that it is unlikely they are mains powered! It is likely that the popularity of EMF meters in ghost research arose simply from the fact that they are easily and cheaply available.

When used to detect electromagnetic pollution, EMF meters produce perfectly useful readings. It is easy to measure a continuous high field produced by nearby electrical equipment, for instance. However, ghost researchers look for unusually high variability and/or 'spikes' in areas of low readings, mostly away from electrical equipment. To do this, they need to understand how EMF meters behave away from the application for which they were designed. This behaviour tends to vary between models.

What causes EMF readings?

When looking at low level EM fields, there are many possible sources of readings. An important one is, of course, electrical equipment and wiring, even if it is not obviously visible. Electrical equipment can be concealed in cupboards or rooms not open to investigators. Electrical cables may be in walls, ceilings or under floors. The fields these cables produce may vary a lot depending on electrical loads from equipment that might be some distance away and not visible to investigators.

Typical frequency response of EMF meterEMF meters detect a low frequency magnetic and/or electric fields over quite a wide range of frequencies (see graph for typical frequency response). Electrical equipment (and wires) gives out both electric and magnetic fields. Any electrically conducting object can block electric fields whereas very little stops magnetic fields. For that reason, magnetic fields tend to be more widespread and stronger than electric fields around electrical equipment. Thus, magnetic fields can often be picked up from relatively remote electrical equipment.

Though electrical equipment and wires produce fields that vary primarily at mains frequency, they can also produce other frequencies at times. For instance, the picture at the top of this page shows the magnetic field frequency spectrum of a washing machine (made using a magnetometer - EMF meters cannot distinguish field frequencies). The low frequency (below mains frequency) peaks are produced by the spinning metal drum. To EMF meters all these different frequencies will simply produce just one overall reading.

As well as a range of low frequencies, EMF meters also typically react strongly to sudden changes in EM fields. Electrical equipment, for instance, can produce such 'spikes' when power is turned on or off or when relays, and certain other components, are activated. In addition, a magnetised object passing close to an EMF meter will typically produce a spike. Even a moving object made out of iron or steel, that is NOT actually magnetised, can produce such a spike. That's because it disturbs the geomagnetic field in the area.

If you are measuring the electric component of a field, as some meters can, you will find it is very sensitive to static electricity. That means that even moving your hand near it is likely to produce a strong reaction. A lesser reaction also occurs even when measuring the magnetic field. That's because electric and magnetic fields are linked and changes in one produce changes in the other (technically, the moving 'static' electric field is interacting with the local geomagnetic field). So, even moving your hand near an EMF meter that is set to detect magnetic fields can produce a spike.

So, even if you do an investigation at a location without mains electricity, your EMF meter can still give readings from natural causes. They might come from moving metallic objects, changing electrostatic fields or all that electrical equipment that is always taken on investigations these days.

Things to try at home with your EMF meter

Before taking your EMF meter to a haunted location, you should get to know the instrument and understand how it behaves in non-haunted places. Here are a few experiments to try at home (in all cases, don't hold the meter but leave in on a solid surface):

  • just leave the meter switched on and watch how the readings vary over minutes and hours, even well away from electrical equipment
  • try moving your hand close to the meter (in both electric and magnetic field modes, if you have them) and look for a reaction
  • wave an unmagnetised piece of iron or steel near the meter and watch the reaction (eg. stainless steel cutlery or a food can)
  • try placing the meter near electrical equipment and watch how the readings change over time
  • see if you can detect the presence of electrical wiring behind walls, near light switches for instance

The frequency problem

A major problem with EMF meters, when used for ghost research, is that they do not discriminate between different frequencies. The readings are only accurate at mains frequency. Readings at other frequencies may be completely wrong (see graph above for typical frequency response). A single reading of '10' might indeed be 10 at 50 Hz but really 20 at 300 Hz. It could also be a combination of different frequency components that add up to '10'. What this means is that, unless you are measuring only mains frequency (which doesn't normally interest ghost researchers) the reading is likely to be wrong.

It is therefore impossible to compare two readings because they may represent completely different frequencies and magnitude even though they both appear as the same figure on the display. Worse, you've no ideas what any of the frequencies are anyway and can't measure them. At least if you knew it was a 50 Hz spike you'd have a good idea it was a field from a mains circuit. In practice, it is impossible to differentiate between a spike of '10' units produced by nearby wires behind a wall and someone walking past with magnetic metal in their pocket, for instance. So to say a field is 'unexplained' simply means what it says - it does not mean you can conclude it is paranormal in origin!

Even if you attach a data logger to a meter you still can't measure the frequency. That's because most meters have a very slow response time (the time taken to react to a change in field) compared to mains frequency. Sampling the reading digitally at a high rate makes no difference to this. Furthermore, EMF meters do not have a flat frequency response so, even if you COULD measure the frequencies, the field strengths would still vary according to frequency!

Field direction

EMF meters generally come in two types - single or triple axis. The relevance of axes is that electromagnetic fields have direction as well as 'strength' (technically, flux density). A triple axis meter is much better because it sums the field in all directions so that if you rotate it, the overall reading should remain the same. A single axis meter, on the other hand, will change reading depending on how you orientate it (without any change in the field you're measuring). Single axis meters will only give a fraction of any EM field they encounter (unless they happen to be pointing in the right direction) making readings even more problematic. It is vital NOT to move or handle single axis meters at all for the whole duration of a vigil. If you do, it will not be possible to compare readings over that period. Also, any slight movement can generate sudden spurious changes in field reading that do not reflect any real physical change at all.

Three phase interference

In large buildings, as opposed to normal houses, where a lot of vigils take place, there is another interesting factor to consider - 3 phase power supply. Most electrical equipment in such a large building will run on just one phase (though some might be connected to all three phases, in a factory, for instance). Though the equipment in different phases is not electrically connected, the different phase EM fields produced may meet in certain locations. Where this happens, it can cause fields to vary more than might be expected, because the fields are out of phase with each other. It might, for instance, cause the field to paradoxically fall when an appliance is turned on. Such phase interference can produce unpredictable changes in field strength. The diagram below (produced by a magnetometer) shows the kind of effect 3 phase interference can have on a magnetic field.

Three phase magnetic effect

Magnetic hallucinations

Laboratory research has shown that certain weak, varying magnetic fields (EIFs) can give some people hallucinations resembling haunting experiences. EMF meters cannot detect such fields directly because (a) they are usually not sensitive enough and (b) they cannot distinguish between frequencies. EIFs might show up on certain EMF meters in special circumstances but you would never be able to differentiate them from readings with other causes. Thus EMF meters have no practical use in detecting EIFs.


It is common practice at the start of some haunting investigations to go around a room looking for particular sources of EM fields and to get an overall 'baseline' reading for the area. The idea is to see if there are spikes above the baseline during the vigil that might correspond with paranormal activity.

The problem with such baseline readings is that they assume that the start of a vigil is somehow typical, average or normal. In fact, EM fields in any building vary all the time (the diagram below shows variation in a building over just 10 minutes), particularly with the use of electrical equipment. To determine an average reading, you'd need to take measurements for at least 24 hours. Even at night there is electrical activity such as lights, pumps, fridges, battery rechargers, etc could all be active. The investigators themselves may be using power for their monitoring equipment. If the 'baseline' reading happens, by chance, to coincide with unusually low readings, many measurements afterwards will appear elevated even if, in reality, they may be typical for that time of day.

Magnetic variation

A single baseline measurement is therefore not very helpful. A much better solution is a 'positional baseline'. With this sort of setup you position a pair of EMF meters so that one is in a 'hot spot' (where lots of paranormal activity has been reported) and the other in a control area (somewhere nearby and as similar as possible but with no reports of activity). In this setup, instead of comparing readings to a more or less random baseline, you compare contemporary readings in different places. The 'control' instrument effectively provides a continuous baseline for the 'hot spot' one. It is then easy to spot if there are higher readings at the hot spot compared to the control.

This sort of setup is much easier if you are using a data logger, feeding readings simultaneously straight into a computer. You can then compare contemporary readings later. Though it is possible to attach some models of EMF meter to data loggers, don't assume that will give you any valid frequency information. EMF meters typically only produce readings a few times a second. So, using a high digital sample rate will not give valid frequency information faster than that. You still won't be able to split out mains frequency, for instance.

Paranormal activity

EMF meters are said to give high readings or spikes when paranormal activity occurs. But how do you define 'paranormal activity' for this kind of exercise? It is important because there is a wide variation for what is accepted as 'activity' between different researchers. Some would say only a full apparition seen by multiple witnesses would count as ghost activity. Others might accept odd noises, feeling cold or hot, seeing strange lights, a feeling of presence or even just feeling a bit odd! Clearly you need to define what is acceptable beforehand, particularly if you are going to compare different haunted locations. In addition, investigations reveal that most paranormal reports actually have xenonormal causes. So how can you be sure that an apparently paranormal event coinciding with a 'spike' in an EMF meter is not really xenonormal instead?

Some researchers accept information from mediums or psychics even when no one else is experiencing anything odd. Because we don't know how such information is obtained by mediums, it is effectively investigating the unknown using a tool which you can't understand! Therefore, it is recommended that you do not include mediumistic data for such work. The same comments apply to such techniques as ouija and seances.

Naturally, all supposed paranormal activity must be thoroughly investigated to see if it has natural causes. Anything that has a mundane explanation must be eliminated from consideration. Natural causes are not always investigated when assumption-led methods are used.


Can EMF meters detect paranormal activity? Given the problems outlined above, it seems unlikely. In summary:

  • EMF meters cannot show frequency data so that different readings cannot be compared
  • the lack of frequency data also makes it difficult to guess the likely source of any reading
  • there are many natural sources of variability and spikes in low field areas (including non electrical sources) which cannot be easily distinguished with EMF meters
  • EMF meters cannot identify 3 phase interference
  • EMF meters cannot identify fields capable of causing hallucinations (EIFs)

In addition, EMF meters are often not used in the best way they could be. Problems include:

  • baselines at the start of vigils which are little better than random
  • what constitutes 'paranormal activity' is often poorly defined making comparisons between different vigils and different sites problematic
  • mediumistic information is sometimes included as 'activity' even when nothing is witnessed by non-mediums
  • paranormal activity is not always investigated exhaustively to eliminate natural causes

The biggest single problem is that if you get a reading on an EMF meter, most of the time you've really no idea what has caused it or whether the figure is correct. You can only really expect accurate figures when using the meter to measure mains-produced fields. The second biggest problem is that you can't rule out EIFs, which could be responsible for any apparent paranormal activity witnessed.

We cannot really conclude anything from the existing vague anecdotal reports. For instance, it is said that EMF meters produce higher or more variable readings at haunted locations. However, it was also said that orbs were more common at haunted location but a systematic study contradicted this claim and other work questionned if they were even paranormal.

Even if a systematic study of EMF readings at haunted locations was done (using the methods recommended above) it would still suffer from the frequency problem. Any higher readings found at haunted locations might actually be caused by EIFs, that are responsible for ghostly hallucinations (though you wouldn't be able to tell!). Also, even if EMF fields spiked predictably as someone was watching an apparition, coincidence would still need to be eliminated. Given the extreme rarity of apparition sighting, it is unlikely that enough examples could be collected to be statistically significant.

The way forward

Using a magnetometer, instead of an EMF meter, to compare hot spots and control areas, would solve many of these problems. You could see the frequency structure of the field and so eliminate mains frequency straight away. You could also look for EIFs and study static fields as well (which may be responsible for ghost reports at Muncaster). Unfortunately, such magnetometers are not cheap.

* Some people claim that electromagnetic fields may produce health problems but is highly controversial and not really relevant to this article

Where did the idea come from?

The idea of ghosts emitting electromagnetic fields seems to have emerged relatively recently. It is tempting to speculate that the idea arose simply because investigators started using the meters, in the same way that the idea of paranormal orbs (which proved to be photographic artifacts) coincided with the early use of digital cameras. Whenever instruments are used at haunted locations, there is inevitably a tendency to attribute unusual readings to the haunting, even without any other corroborating evidence.

A more speculative notion is that the idea may have been prompted, in some way, by Persinger's laboratory work that suggested that some ghostly experiences might be magnetically-induced hallucinations (EIFs - experience inducing fields).

© Maurice Townsend 2008, 2011